ABOUT THIS SERIES For the past 10 months, Chronicle staff writers Reynolds Holding and William Carlsen investigated allegations of widespread legal and ethical misconduct in Silicon Valley. This five-part series is based upon thousands of pages of financial reports, internal company memorandums, government records and court documents, as well as scores of interviews with corporate executives, government officials, state and federal prosecutors, judges, lawyers, academicians and entrepreneurs. Some of those court documents were unsealed only as a result of a court order obtained by The Chronicle in San Mateo Superior Court. PHANTOM RICHES Beneath the glitter of booming Silicon Valley, executives have been accused of lying about their products and doctoring their books, leaving devastated investors in their wake FIRST OF FIVE PARTS Silicon Valley is the epicenter for the fastest creation of wealth in history. But the high-tech miracle has a dark side: Untold stories of ruined investors, betrayed entrepreneurs and regulators who are overmatched and overwhelmed. In 1989, Lev Dawson scoured Stanford University's Jackson Library, searching for a shot at Silicon Valley gold. He found it in an unproven idea for making laptop and cell phone batteries smaller, lighter and longer-lasting. With venture capital backing, Dawson created Valence Technology Inc., and during the next four years the batterymaker issued more than $167 million in stock and signed potentially lucrative contracts with Motorola and Hewlett-Packard. Unfortunately, the batteries had a few problems. They sparked and exploded and burned. One burst into flames so intense that it melted through the factory's tile floor and into the concrete foundation. A senior company engineer said carrying a battery was like having ``a bomb in your pocket.'' But word of the batteries' failures would not get out for several years. By then, Dawson was farming sweet potatoes in northeast Louisiana, having unloaded more than 2.4 million shares of his Valence stock for almost $41 million.

Public stockholders, though, were almost wiped out. Relying on company assurances that all was well, they bid up the stock to more than $25 a share before the truth emerged in 1994, and the price plunged to $3.25. Jim Varano, a 53-year-old pharmacist living near Boston, believed Valence's prospects were so promising that he borrowed money for a $320,000 stake in the company. After his investment evaporated, he had trouble sleeping, working and paying his sons' tuition. Today, he minces no words: ``They're crooks and swindlers.'' In recent years, Varano and tens of thousands of investors like him have filed lawsuits accusing Dawson and other Silicon Valley executives of securities fraud, insider trading and legal and ethical misconduct. The executives -- including Dawson -- deny any wrongdoing. But officials at scores of high-tech companies, from bedrock hardware manufacturers to high-flying dot.coms, have been accused of lying about revenues and products to satisfy market expectations, then dumping their stock for millions of dollars in illegal profits. At the same time, lawyers at the valley's most influential firms have been accused of playing fast and loose with conflict-of-interest rules, in some cases allegedly betraying the clients they were committed to represent. And some of the industry's most prominent venture capitalists, the financiers of the high-tech boom, are alleged to have stolen ideas and companies from the entrepreneurs they financed. But the U.S. Justice Department and the Securities and Exchange Commission have largely failed to enforce laws enacted to protect the markets' integrity. During the past decade, the agencies have filed only a few criminal and civil cases in Silicon Valley, despite warnings that corporate misconduct is thriving. ``We've seen a dramatic increase in the number of complaints and investigations -- far more than three years ago. That's got to reflect a big increase in wrongdoing,'' said David Bayless, until June the head of the SEC office in San Francisco. ``And it's the same goddamn thing every time: faked revenues, 35-day months, side agreements, hiding things in warehouses -- you name it.'' This is a series about the dark side of the Silicon Valley miracle. It is about some of the most egregious cases of securities fraud, trade-secret theft and abuse of small entrepreneurs -- cases largely concealed from public scrutiny by nondisclosure agreements, sealed court documents and fears of retribution. Set amid the tidy light-industrial parks and sprawling corporate campuses that line the south end of San Francisco Bay, it is a story about greed and hubris, about high-powered executives and boardroom dealmakers who operate with flagrant contempt for the rules of law and ethics.

The story begins with two men -- a plaintiffs' attorney and a high-tech pioneer -- whose battles with each other reflect the fundamental conflict between unrestrained corporate freedom and the need to protect the rights of investors. A SON OF THE GREAT CRASH, 1946 William Shannon Lerach -- the attorney despised most by corporate America -- was born a victim of the stock market. In September 1929, after receiving a substantial inheritance, Lerach's father, Richard, went to work as a stockbroker. He invested everything he owned in the market. The next month, the market crashed, and the elder Lerach was ruined. In 1933 and 1934, President Roosevelt signed landmark securities legislation to prevent the kind of speculation, manipulation and outright fraud that triggered the crash and caused millions of Americans to lose their savings and jobs. The legislation came too late for Richard Lerach, who was relegated to selling metal parts for the rest of his days. But for his youngest son, William, it would provide the means for a career of wealth and notoriety, one that would ultimately earn him the sobriquet ``prince of darkness'' -- and a law firm draw last year of more than $16 million. Bill Lerach was born in 1946 in the Ohio River Valley. As he grew up, he worked in a series of dead-end jobs at a plant nursery, a funeral home and a pool hall. He was a mediocre student until law school at the University of Pittsburgh, where he graduated second in his class and attracted a job offer from Pittsburgh's elite firm, Reed Smith Shaw & McClay. He excelled there, making partner in less time than any other lawyer in the firm's history. He was brash and profane and wickedly funny, a flush-faced guy with a big head of puffy red hair. But he was never entirely comfortable representing companies in lawsuits and deals. He could never overcome the deep mistrust of corporate America he had developed while watching his father work for almost nothing. ``They treated him like they treat pencils,'' he said, ``shaving him down until there was nothing left but a nub.'' The last straw was a lawsuit filed against his corporate client by a shareholder. ``This guy's case was so good, and he was so right,'' said Lerach, ``and we just chopped him up into little pieces. We just blew him away. And I thought, if that guy had had more money and a better lawyer, he would have won.'' Lerach realized he wanted to represent shareholders -- not to mention earn bigger paydays. So in 1976, he took a job with prominent plaintiffs' attorney Melvyn Weiss, opening a one-lawyer branch office in San Diego for Weiss' New York firm, specializing in suing companies for securities fraud.

The office prospered and quickly expanded, taking cases around the country. But it did not come into its own until 1984, when, for the first of several times, it filed a lawsuit against a Silicon Valley company called Seagate Technology, run by a man named Alan Shugart. THE DISK DRIVE REVOLUTION, 1979 Al Shugart was in a hurry from the start, impatient to leave behind a downscale adolescence spent surfing and fixing motors near Chino, in San Bernardino County. In 1951, the day after earning an engineering degree from a private college called University of Redlands, he landed work as a repair technician for International Business Machines. Promotions came quickly. At the company's research center in San Jose, he led a team of top-flight engineers credited with developing the first rigid disk drives, a revolutionary means of storing data. It was an extraordinary time in Silicon Valley, the eve of the computer revolution. And in 1969, when IBM wanted Shugart to join its New York headquarters, he quit instead, ending 18 years with the company. He worked for Memorex for several years, then started his own floppy disk drive business -- Shugart Associates. But his role in the enterprise ended badly in 1976 when the directors of his company fired him for reasons he cannot -- or will not -- explain. For the next three years, Shugart fished for salmon and tended bar in Santa Cruz. Then, in 1979, he hooked up with Finis Conner, a former colleague from Shugart Associates. Demand for personal computers was rising fast, and Shugart and Conner were determined to catch the wave. Their idea: a hard disk drive with five times the capacity of a standard floppy disk -- at half the price. The two men recruited three more engineers through newspaper ads and then created the company that would become Seagate Technology. Seagate's new disk drives began to roll off production lines in 1980. It was an increasingly competitive market. Apple Computer, IBM and other major customers were starting to make their own hard drives. Japanese manufacturers were jumping in with even cheaper products. Still, by the summer of 1984, Seagate claimed profits of $42 million, up 221 percent over the previous year. Shugart and other Seagate executives crowed about the company's bright prospects for growth, sending the stock price -- and Shugart's reputation as a Silicon Valley entrepreneur -- soaring. Later that summer, though, the Seagate success story began to unravel.

In August, without warning, Seagate executives announced that they expected dramatically lower profits and slower growth. The company would have to set aside millions of dollars to cover obsolete inventory and accounts that might never be collected. The huge Watsonville plant -- touted as a symbol of the company's astounding growth -- had, in fact, never been used, and it was for sale. The earnings report that Seagate issued in late 1984 showed precipitous declines in revenues and profits. Almost instantly, Seagate stock plummeted to $4.38 a share -- a 75 percent drop from its high of $17.13 in June 1983. Public shareholders took a beating. But shortly before the company issued its dire report, Shugart and other executives had unloaded much of their own stock for gross proceeds of more than $84 million. Shugart's take alone was more than $15 million. Bill Lerach smelled a rat. THE LAWYER DOLL, 1984 Lerach had been filing class-action lawsuits on behalf of shareholders for years, extracting multimillion-dollar settlements from notable companies such as Gap Inc. and Memorex and -- with the lawyers' standard one-third contingency fee -- amassing a personal fortune. But Shugart and Seagate were prized game, potential trophies from the heart of Silicon Valley. And less than a month after the company released its financial report, Lerach and several other lawyers filed class-action suits that accused Shugart and his colleagues of securities fraud and insider trading. In Seagate's Scotts Valley headquarters, Shugart was furious. He vehemently denied the allegations and vowed to fight the suits no matter how long it took, no matter how much it cost. He hated anyone who tried to obstruct his entrepreneurial freedom. He hated lawyers most of all. ``Al Shugart hated lawyers so much he kept a lawyer doll on his desk so he could periodically snap its head off,'' said one plaintiffs' attorney. ``But he really detested Bill Lerach.'' LERACH'S WRATH, 1986 In 1986, as the Seagate case dragged through its second year of pretrial motions, Lerach turned his attention to a lawsuit that had the potential to make him the nation's best-known class-action attorney. It involved Nucorp Energy Inc., a San Diego oil company that issued $150 million in securities in 1981 before declaring bankruptcy a year later. Lerach and the SEC contended that the company used phony revenues to lure investors into a disastrous deal.

Lerach tried the case to a jury verdict, his first in more than 18 years of practicing law. He was so sure of victory that he told the lawyer on the other side, ``This is going to be an ignominious end to your mediocre career.'' When jurors requested a calculator, he asked the judge to make sure it could handle nine figures. On April 25, 1988, he invited his colleagues and the media to court for the verdict. Then he lost. Lerach was stunned. His face blanched, then burned a furious red before he stormed from the courtroom. His colleagues say the fury lasted weeks. But rather than weaken Lerach, the humiliating defeat seemed to galvanize him for battle against the titans of corporate America. It intensified his belief, as he wrote in a memo about jurors one week later, that ``60-year-old Republicans'' were inherently unsympathetic toward victims of securities fraud, and a person's ``position in society, his sense of class, is probably the single most important factor in determining his attitude toward (investors' claims).'' And Shugart -- 58 years old and rich -- would soon experience the force of Lerach's wrath. THE IBM DEAL, 1988 By the end of 1987, Seagate had recovered from its financial problems to become one of the most prosperous companies in the high-tech industry. And Shugart, undaunted by the lawsuit Lerach had filed against him, emerged as one of the most visible and outspoken executives in Silicon Valley. With a new 3.5-inch disk drive, Seagate earned revenues surpassing $1 billion a year. The company's sprawling new production facilities in Thailand and Singapore made it the largest private employer in both of those countries. And in March 1988, Shugart announced a new order from Seagate's biggest customer -- IBM -- a major announcement that nervous investors had long been waiting for. Wall Street applauded, and the company's stock price soared, but there was a problem: The deal didn't require IBM to order anything from Seagate. IBM executives were horrified by Shugart's announcement. They called him and another Seagate executive to Washington, D.C., in May. Over dinner, the men from IBM accused Shugart of blatantly violating a written agreement that barred him from announcing the deal.

During a heated exchange after dessert, the ranking IBM executive, Irv Abzug, accused Shugart of trying to please ``security analysts'' and ``hype'' Seagate's stock. At that point Shugart walked out. For more than a month, he assured financial analysts that demand for Seagate's disk drives was so strong that the company would expand production and even raise prices -- an unheard of move in the highly competitive disk-drive industry. But in July, Seagate disclosed that its earnings were down 57 percent from the previous year. Over the next day, the value of Seagate shares plunged $250 million -- a record for high-tech stocks at the time. Shugart, however, had unloaded 100,000 shares in April, when the stock price was up. Lerach struck quickly. On July 29, 1988, just three months after the devastating Nucorp verdict, he filed a second class-action suit against Seagate, again accusing Shugart and other top executives of securities fraud and insider trading. In a rage, Shugart categorically denied the allegations -- and vowed to fight them with everything he had. THE TRUCKER'S TIP, 1989 A tip came late the next year. Two attorneys working with Lerach on the Seagate case flew to San Jose and knocked on the door of a trucker named George Armour, who had been employed by the shipping firm that Seagate used in 1983 and 1984. For four years Lerach and his associates had heard the rumors that Seagate boosted revenues by claiming to have sold disk drives that were actually stored in warehouses. And now, it seemed, the rumors could be true. At first Amour was reticent, but then he told the attorneys how Seagate cleared disk drives out of its warehouses at the end of every business quarter. He told them that he and other truck drivers often worked 16- to 18-hour shifts, hauling semitrailer trucks packed with disk drives to rented warehouses in Santa Clara and San Jose. At one point, he said, the company parked at least six fully loaded trailers in the Santa's Village parking lot in Scotts Valley. The trailers were parked back-to-back, so the doors could not be opened. A security guard patrolled the lot. ``It seemed a little funny to me at the time,'' Armour said. ``Those disk drives cost a lot. They were like gold.'' Later, a second trucking employee, Peter Page, described how semitrailer trucks loaded with pallets of Seagate disk drives arrived through the night at a local freight

warehouse. Two trucking lines delivered the merchandise at the end of a financial quarter in 1983 or 1984, he said, and it was not removed until later the next month. The truckers' statements would be the turning point in Lerach's first suit against Seagate. He would eventually force Shugart to the bargaining table for a settlement: Without admitting any wrongdoing, Seagate agreed to pay $5 million to 17,000 shareholders. Lerach and the other attorneys who had sued the company collected $3.1 million in fees and $1.26 million in expenses. A CALL TO ARMS, 1990 ``Enough is enough,'' blared the full-page ad from the back cover of the April 1990 Upside magazine, a glossy monthly that covers the high-tech industry. ``Are securities lawyers holding your company ransom?'' The ad was Shugart's, a public declaration of war against class-action lawsuits. For years, scores of executives at some of the most prominent companies in Silicon Valley -- including Oracle, Apple and Sun Microsystems -- had been sued for securities fraud. And whatever the merits of the allegations, legal fees and settlements cost the high-tech industry millions of dollars every year. In the ad, Shugart asked other executives to send him their business cards if they wanted to help stop the lawsuits against their companies. His plans were uncertain, but during the next few weeks, he received a couple of dozen business cards. Then a mysterious envelope arrived from San Diego. In it was one of Lerach's business cards. Scrawled on the front was a message: ``Dear Al: More is coming.'' True to his word, on June 25, 1991 -- just 11 days after the settlement of his first suit against Seagate – Lerach struck again, filing a third lawsuit accusing Seagate and Shugart of securities fraud. EXPLOSIONS IN THE LAB, 1991 The class-action business was making Lerach rich, providing him the means to support a sprawling mansion in Rancho Santa Fe, an enclave of multimillion-dollar homes in the Southern California hills overlooking the Pacific. It was also earning him the enmity of corporate America, whose executives called him ``pond scum'' and ``blood-sucking scumbag'' while commiserating with other targets of shareholder suits about being ``Lerached.''

But no one despised Lerach more than Shugart, who in 1994 found himself a defendant in yet another shareholder suit -- this one over his involvement with Valence Technology. Shugart had been asked to sit on Valence's board of directors by Carl Reed, Seagate's former vice president of operations. A tough and efficient manager, Reed was in many ways responsible for Seagate's rise to a billion-dollar corporation with almost 40,000 employees around the world. In May 1991, Reed left Seagate to become president and chief operating officer of Valence in San Jose. Several months later, he persuaded Shugart to join the company's board. In theory, Valence's technology was compelling: a wafer-thin lithium polymer battery that would last four times longer than standard batteries, would cost one-tenth as much to make -- and would contain no liquids that might cause the lithium to explode. From the beginning, though, Valence had problems. According to former Valence engineers, virtually all of the thousands of machine-made batteries eventually shorted out, emitting sparks or intense heat when pinched by assembly equipment. The equipment itself was often broken, they say, hampering attempts to show off the factory to potential customers or investors. When visitors arrived, employees would pretend the machines were working by sending hand-made batteries down the assembly line. And the cost of manufacturing was prohibitive: between $50 and $60 for a battery the company planned to sell for about $15. ``It was like strapping a $100 bill on one of these things and then sending it out,'' a former manager said. But Chief Executive Officer Lev Dawson, Reed and other top executives publicly claimed that Valence was on the verge of success. In May 1992, the company said basic research on the state-of-the-art battery was complete. In December, Valence announced a $100 million order from Motorola to supply batteries for cell phones and pagers, and it set early 1994 as the delivery target date. Then, in February 1993, an article in Forbes magazine questioned the company's prospects. But Dawson controlled the potential damage, telling `̀ Valence Followers'' in an open letter, ``We remain in solid financial position.'' A few days later, Dawson stepped down as CEO. Reed replaced him, and the optimistic announcements continued.

In late 1993, Valence touted plans to open a huge new plant in Ireland. And in January 1994, the company announced a $3.5 million contract to supply Hewlett-Packard with batteries for laptop computers. ``We look forward to producing our first commercial batteries in 1994,'' Reed said in a press release. BAD NEWS REACHES BOSTON, 1994 Across the country, in Braintree, Mass., Jim Varano read about the developments with excitement. He already owned a few hundred shares of Valence, and the latest news persuaded him to buy even more. ``In effect, they were saying that the batteries were perfected,'' he said. ``They had delivery dates. They were going to build a production plant. It meant they had a product ready.'' He talked it over with his broker, and in early 1994 purchased approximately 25,000 shares on credit. Meanwhile, unknown to Varano, Dawson and other Valence insiders were unloading millions of shares of their own. On May 4, 1994, on his way to work at the Veterans Affairs hospital in nearby Brockton, Varano stopped at a Dunkin' Donuts for a cup of coffee. He sat down and turned to the business section of the Boston Globe, eager to see whether Valence stock had continued to climb. To his horror, he saw that the stock had plunged almost 50 percent. Valence had announced it could not meet Motorola's requirements for batteries -- and probably never would. Overnight, Varano lost hundreds of thousands of dollars. ``I was devastated,'' he said. Even worse news followed. In June, the company announced that it could not deliver batteries to Hewlett-Packard. And in August, it announced that it would abandon development of a solid lithium battery. Almost immediately after the stock crashed, Lerach and other plaintiffs' attorneys filed class-action lawsuits accusing Valence and Dawson, Reed, Shugart and other company executives and directors of securities fraud. Varano became the lead plaintiff in Lerach's suit. He has yet to recover a dime. THE REFORM ACT, 1995 On Dec. 15, 1995, Lerach went to dinner at the White House.

At the time, President Clinton was grappling with the Private Securities Litigation Reform Act of 1995, far-reaching legislation designed to limit securities-fraud class actions against corporate executives. It proposed the most dramatic change in the nation's securities laws since 1934, and Clinton had four days to sign or veto the bill. Lobbying for the Reform Act had started in earnest two years earlier, when accounting firms and Wall Street investment banks complained to Congress that they were being dragged into securities suits filed against their corporate clients simply because they had ``deep pockets.'' Their plea for relief went nowhere at first. Then, in November 1994, the Republican Party swept into Washington brandishing the ``Contract With America,'' which included a provision against securities class actions. In just two months, the new Republican-led House of Representatives passed strict rules designed to put plaintiffs' attorneys out of business. The most draconian sections, which would have eliminated almost all shareholder suits, were weakened in negotiations with the Senate. Top Silicon Valley executives, in their first serious display of political power, lobbied the president to sign the Reform Act. They argued that most securities fraud suits were frivolous. As proof, they pointed out that half of the 150 largest companies in Silicon Valley had been the focus of those lawsuits, implying that securities fraud could not possibly be that widespread. But the Democratic Party's traditional supporters -- consumer groups, civil libertarians and plaintiffs' attorneys such as Lerach -- pushed the president for a veto. Lerach was one of Clinton's earliest and strongest political backers. During the 1992 presidential campaign, he rode on Clinton's bus through the nation's heartland. He was host for a fund-raiser for Clinton at his mansion in Rancho Santa Fe. Lerach disputes published reports that suggest he persuaded Clinton at the December White House dinner to veto the Reform Act. ``I was at a dinner with President Clinton and 500 other people,'' he said. ``I shook his hand and never spoke to him about the bill. There were 20 lobbyists on the other side who were at that dinner.'' But four days later, 20 minutes before the midnight deadline, Clinton vetoed the Reform Act. ``The president supports the goals of this legislation,'' said spokesman David Johnson in announcing the veto, ``but he is unwilling to sign legislation that would have the effect of closing the courthouse door on investors who have legitimate claims.'' The veto would not last. For the first time in Clinton's presidency, the House and Senate voted to override, and the Reform Act became law.

SHUGART'S REVENGE, 1995 Corporate executives around the nation celebrated, and nowhere was the news greeted with more joy than in Silicon Valley. Shugart was ecstatic. Four times he had been sued by Lerach. The first case had cost him millions. The second had been thrown out of court. The third would be settled for millions more. He had been dropped from the Valence suit because he had not participated in running the company. Finally, it was payback time. He fired off a fax to Lerach's San Diego office. It said: ``Dear Bill, More is coming. Sincerely, Al.'' SEAGATE TECHNOLOGY With headquarters in Scotts Valley, Seagate is the world's largest manufacturer of computer disk drives. Seagate was founded in 1979 by Alan Shugart and Finis Conner. Shugart, who is credited with leading the team of IBM engineers that invented the floppy disk, served as the company's CEO until July 1998. MILBERG WEISS BERSHAD HYNES & LERACH Milberg Weiss is the largest law firm in the nation to focus on class-action lawsuits. The firm specializes in securities fraud suits and has filed hundreds of shareholder suits against Silicon Valley companies, resulting in more than $2 billion in settlements. William Lerach joined the firm in 1976. He has successfully sued some of the most notorious figures in two of the nation's biggest financial scandals, winning a $250 million settlement from Charles Keating and American Continental Corp. and $1.5 billion from Michael Milken and Drexel Burham Lambert. Silicon Valley is the epicenter of the fastest creation of wealth in history. But the high-tech miracle has a dark side: untold stories of ruined investors, betrayed entrepreneurs and regulators who are overmatched and overwhelmed. Robert Crowe was afraid he wouldn't be ready.

For months, the assistant U.S. attorney had been preparing a securities fraud case against executives at California Micro Devices, a Milpitas chip manufacturer. But as the 1998 trial neared, Crowe was inundated with 600,000 pages of documents and didn't even have a clerk to work the copy machine. Meanwhile, the FBI and the Securities and Exchange Commission continued to bring him promising cases against other Silicon Valley companies. But Crowe was the only federal prosecutor in San Francisco working full time on investment fraud, and there was nothing he could do with them. ``The inventory of cases kept building,'' said Crowe. ``I couldn't keep up.'' Cal Micro was the first major Silicon Valley fraud case tried by the U.S. attorney's office in San Francisco. It was also the last -- even though shareholders have sued more than 60 high-tech companies in Northern California's federal courts since 1995. Silicon Valley executives contend that most of the complaints are frivolous, but Crowe and others say the rising number of class-action suits shows that fraud is rampant in the high-tech industry. ``If they put together a special federal task force and sent it into the valley, they could bring a ton of fraud cases,'' said one senior prosecutor at the U.S. Department of Justice. But the U.S. attorney's office has done little to deter securities fraud in Silicon Valley, filing criminal charges against only a handful of high-tech executives the entire decade. Since shaking the recession of the early 1990s, Silicon Valley has become the epicenter of the fastest creation of wealth the world has ever seen. And the opportunities and motives for financial deceit have never been greater. Silicon Valley executives face intense pressure to meet Wall Street's expectations. Missing quarterly projections can mean financial disaster for a company -- and for the executives whose compensation packages depend on the company's stock price. For many corporate officers, the temptation to ``cook the books'' -- and then unload shares before the company's stock price collapses -- is overwhelming. And unless federal prosecutors crack down, experts warn, securities fraud will continue to flourish. ``More executives should be going to jail,'' said Stanford University law Professor Joseph Grundfest, a former SEC commissioner and national expert in securities fraud. ``That will grab their attention . . . and hand a valuable lesson to the entire economy.'' This is the story of how federal prosecutors have allowed securities fraud to spread through Silicon Valley like a virus. It is an account of the one major case that the U.S. attorney's office in San Francisco prosecuted, desperately hoping the trial would serve as a warning to the high-tech industry.

And it is about an even bigger case, one of the most flagrant in Silicon Valley history, that federal prosecutors let sit on the shelf. Yamaguchi's Vow, 1993 When Michael Yamaguchi was appointed U.S. attorney for the Northern District of California in 1993, he took over an office that had been adrift for years under a string of temporary chiefs. As one of his first priorities, Yamaguchi pledged an aggressive campaign against white-collar crime. But the soft-spoken tax specialist quickly learned it wasn't going to be easy. ``We did not have nearly enough resources,'' said Richard Seeborg, a former prosecutor in the agency's San Jose office. ``When you compare our office to U.S. attorney's offices around the country, we were grossly understaffed.'' In the 1990s, the Northern California district office had, on average, only one-quarter of the federal prosecutors per capita that were assigned to lower Manhattan and one-third the number of prosecutors posted to Southern California. Yamaguchi's first test came in June 1994, when former Chronicle columnist Herb Greenberg reported allegations from former employees at California Micro Devices that the company had been booking fake or premature revenue for years. Shortly after the disclosures, as the FBI and SEC investigated, the company's senior financial officer, Steven Henke, resigned. A few weeks later, the board of directors fired Chief Executive Officer Chan Desaigoudar, and Cal Micro's stock plummeted an estimated $139 million. As Yamaguchi's prosecutors conducted interviews and pored over Cal Micro's records, they found that the company had booked millions of dollars in revenue for products that had not been shipped. They also discovered double bookkeeping, false financial filings and illegal write-offs. At the same time, the prosecutors discovered, Desaigoudar and Henke had dumped more than $1.1 million of their own stock before its price collapsed. It was the perfect case for Yamaguchi to demonstrate that his office was serious about white-collar crime. The Mastermind, 1989 Phil White was always ambitious -- even before he was accused of masterminding one of the largest securities fraud scandals in Silicon Valley. The sandy-haired White grew up in small-town Illinois, the son of an accountant and a public school teacher.

He majored in business at Illinois Wesleyan University and later graduated from the University of Illinois' business school. He first worked for a travel company, leading tours through Latin America, Europe and Canada and running the company's Hawaii office. But when he realized he had no chance to head the family-owned business, he quit. ``I didn't go to school,'' he once told a financial trade magazine, ``to be second or third fiddle.'' Next, he moved to IBM's St. Louis office, where he consistently won sales awards. But he knew he would never challenge for the top job at IBM, either, and after 15 years, nearing middle age, he decided to change course. In 1982, White moved to a sales and marketing job in Silicon Valley and two years later joined the board of Wyse Technology. In 1986, when the company needed an aggressive new president and chief operating officer, White seized the chance. Wyse had been a prosperous manufacturer of video display terminals for many years, but its market niche was doomed by a new generation of desktop computers that could operate without mainframes. Shortly after White took over, Wyse moved into the personal computer market, and White's management team promised 20 percent revenue increases every three months. It was an ambitious goal, and almost immediately it led to disaster. In the haste to meet the target, Wyse made a lot of shoddy computers. They overheated, erased information and ejected circuit boards during shipping. The pressure to meet revenue goals grew so intense that in December 1987 the company shipped computers in garbage bags after the anti-static containers ran out. According to court documents, the company claimed sales on computers that were actually sent ``around the corner'' to shippers' warehouses, then returned to Wyse after the end of the financial quarter. Some computers were not shipped at all, merely entered into records as ordered and ``processed.'' On Feb. 16, 1988, the company reported quarterly revenue of $128 million -- an astonishing 75 percent increase over the quarter one year earlier. At an industry conference in September, White proclaimed that 1988 ``looks to be a banner year for all of us.'' But on Jan. 5, 1989, the company abruptly announced that revenues for the previous quarter were down by half from a year earlier. The stock price plunged to $5 a share, far below its high of $25.50. Almost 600 employees -- 15 percent of the company's workforce -- would soon be fired. Before the year was over, a Taiwanese company would buy Wyse for $10 a share.

Investors, mutual funds and pension funds lost tens of millions of dollars. Bill Lerach and other plaintiffs' attorneys sued White and other company officials for securities fraud and insider trading. The suit was settled in 1992 for $15.5 million in cash -- with no admission of wrongdoing. It was not the last time White would be accused of ``cooking the books.'' THE PROSECUTOR, 1990 Robert Crowe charged out of Cornell University law school in 1983, eager to become a criminal defense lawyer. But he quickly found positions as a public defender more difficult to find than prosecutor jobs, so he wound up as an assistant district attorney in Brooklyn, N.Y., and loved it. After several years of prosecuting street crimes, the Chicago native began looking for something more challengng. He found it in 1989 when he went to work in the U.S. attorney's branch office in San Jose. In 1990 he got his first securities fraud case. It involved a company in Campbell called StarSignal Inc. At the time, the U.S attorney's office didn't pay much attention to investment-fraud cases. According to several former prosecutors, the lawyer responsible for them just sat on the complaints in the San Francisco office. It got so bad, they said, that frustrated SEC officials stopped sending cases over. ``The head of SEC enforcement was thrilled when I began working on StarSignal,'' Crowe said. StarSignal was the brainchild of an excitable and brilliant engineer named Robert Widergren. With several million dollars raised from private investors, the company developed the world's first commercial color fax machine. But the early machines were expensive, costing as much as $26,000, and sales never took off. In 1990, a former executive tipped the SEC that the company was raising money by sending investors false information -- including news of an imminent sale in Spain worth $83 million. The figure, said the tipster, was ``extracted from the air.'' That August, the FBI moved in, arresting Widergren after learning that he planned to transfer money to Belize in Central America. Widergren was charged with bilking investors of more than $3 million. The StarSignal case gave Crowe his first inkling that securities fraud was more widespread in Silicon Valley than anyone suspected.

``Why are you doing this to me?'' Crowe remembers Widergren asking during breaks in the 1991 StarSignal trial. ``My case is just a few million dollars. There's fraud going on all over the valley worth hundreds of millions.'' INFORMIX'S TURNAROUND, 1993 For nearly a decade, Informix flourished as one of Silicon Valley's leading makers of software for managing computer databases. But in 1989 the Menlo Park company bought a software firm in Kansas City, and the acquisition started to drag revenues down. So Informix founder Roger Sippl turned to White, unfazed by Wyse's collapse and the allegations of securities fraud. White quickly swung into action, firing a fifth of the company's workforce. He expanded the company's business to Europe and Asia. He directed Informix's engineers to change the database software so it would work in networks of personal computers rather than just mainframes. When PC networks became the rage, Informix cleaned up. Within four years, the company rebounded from a $46.3 million annual loss to post a $56 million profit. Its stock rose from 56 cents a share to more than $30. The numbers dazzled Wall Street, and White looked like a genius. But then the industry began to slow. And the next year, company executives found new ways to keep revenues growing. THE FIRST WARNING, 1994 Something was wrong with the numbers. In May 1994, internal auditors at Informix were examining the Australian accounts, and the figures didn't add up. The company's Australian subsidiary had booked sales a full quarter before the software products were even shipped. A few months later, the company's outside auditors -- Ernst & Young -- warned company officers of similar problems in Europe. The auditors also chided Informix for selling software in Latin America on ``handshake'' agreements. ``Shareholders will expect agreements with customers to be documented,'' the auditors wrote in a draft memo for the 1994 year-end audit. Informix executives dismissed the incidents as aberrations.

But at an annual meeting in January 1995, Chief Financial Officer Howard Graham met with sales representatives to backdate contracts so that January deals appeared to have closed in December, according to a shareholders' suit. The result was inflated revenues for 1994. The executives joked among themselves that they were closing the quarter on ``December 45th.'' A COMPANY ON A ROLL, 1996 In early 1996, White negotiated Informix's $400 million purchase of Oakland-based Illustra Information Technologies. It seemed a stiff price for a company with less than $10 million in annual sales and an unproven technology that stored images and sound in electronic databases. But White touted the acquisition and the wonders of Illustra's technology. ``This stuff is going to change the way people think,'' he said. And Informix, it seemed, was on a roll. White declared that he expected the firm ``to become a billion-dollar company in 1996.'' But for nearly two years, auditors had warned Informix's board of directors that White, Graham and other executives were inflating revenues through a variety of questionable accounting practices, including backdated sales, ``burn deals,'' barter transactions and side letters. Under what Informix employees called ``Howard's rule,'' Graham would allow sales contracts to count toward revenue even before they were fully signed. In barter transactions or ``Phil deals'' -- named for Phil White -- Informix would agree to buy another company's products if that company would license Informix's software. Informix would then count as revenue the entire value of the software licenses -- without disclosing that it still had to buy the other products before it could get paid. In some cases, it never received payment because the software was returned. And in complex ``burn'' transactions, Informix would agree to sell software licenses to a distributor, then loan the distributor money for the purchase price. Informix would book the transaction as revenue and the distributor would, in theory, resell the software so that it could repay -- or ``burn'' -- its loan commitment to Informix. But in many cases, the distributor was under no obligation to resell the software or to repay the loan. Sometimes Informix would never even deliver the software. The sales, though, remained on the company's books.

In other transactions, White would negotiate side letters that granted customers undisclosed concessions such as price breaks or longer payment terms. The letters changed the deal so that the original sales contracts -- and the revenues recorded under them -- were no longer accurate. As Ernst & Young was privately warning Informix directors about these practices, though, it was publicly issuing letters giving the company a clean bill of health. On April 15, 1996, Informix reported $204 million in revenue for the first three months of the year -- a 38 percent jump from the year before. Wall Street was impressed. The next day, Informix stock rose $5 to $24.25 a share. Over the next three weeks it would top $26. And White and Graham would begin to unload tens of thousands of their shares in the company. DUMPING STOCK, 1996 Auditors could barely conceal their frustration about the accounting irregularities at Informix. ``It just gets better all the time,'' wrote one Ernst & Young auditor after discovering several improper deals. In 1996, Informix booked tens of millions of dollars from transactions that were backdated, incomplete, subject to secret conditions -- or simply nonexistent, court documents show. ``Phil deals,'' or barter transactions, accounted for much of the improperly recorded revenue. That year, Informix sold about $170 million worth of software to companies from which it bought about $130 million worth of computer equipment. Although Informix denied the transactions were related, an audit would later determine that at least some were, and that they had artificially boosted revenues by $55 million. Burn deals and side letters contributed tens of millions of dollars more. By mid-1996, Informix's European subsidiary had booked more than $105 million from burn deals for which the company remained unpaid. The tactics served their immediate purpose: They increased revenues and kept the share price high. By December, before leaving to become chief financial officer at Siebel Systems, a San Mateo software company, Graham had sold more than $2.8 million worth of his Informix stock. During the next two months, White and other executives dumped more than $20 million in shares.

Then, suddenly, everything began to come apart. INFORMIX IN FREE FALL, 1997 Despite its inflated revenues and soaring stock price, Informix was starved for cash. And as the company tried to collect on the burn deals it had written, customers objected, and, in some cases, refused to conduct further business with the company. British software firm Logical Systems International, for example, had agreed to ``buy'' Informix software with the understanding that Informix would resell it and not make Logical Systems pay. ``Informix asked us to do them a favor,'' wrote managing director Stewart Ashton in a March 27, 1997, letter to the company, a favor that would ``artificially inflate the quarter numbers (of Informix) for last June and September.'' But when Informix squeezed Ashton for payment, he howled at having been ``coerced into this . . . arrangement.'' Finally, Informix could no longer hide its crumbling finances. On May 14, in its quarterly filing with the Securities and Exchange Commission, Informix announced that it had lost $140.1 million in the first quarter of 1997, and that ``almost half of the licenses sold to resellers'' since 1995 ``have not been resold.'' White blamed an ``overemphasis'' on selling the new Illustra-based technology rather than traditional software, but financial analysts knew better. ``An unmitigated disaster,'' one analyst called the company's announcement. Yamaguchi's Retreat, 1997 Robert Crowe watched from his desk in San Francisco as much of the U.S. attorney's office sank slowly into disarray. He had transferred from the San Jose office after Widergren's conviction in the StarSignal case. He had taken charge of screening most investment-fraud prosecutions and coordinating investigations with the SEC. But Yamaguchi's vow to crack down on white-collar crime in Silicon Valley quickly turned hollow. In December 1996, Sen. Diane Feinstein recommended Yamaguchi to fill a vacancy on the U.S. District Court in San Jose. The judgeship should have been a crowning achievement for the 46-

year-old prosecutor, whose father, grandparents and thousands of other Japanese Americans were kept in World War II internment camps after a series of rulings by federal judges. But five months later, Yamaguchi withdrew his name after his public comment about an important drug prosecution led to a mistrial in the case. ``After that, and all the bad press around the case, Mike just disengaged and hid out in his office,'' said one former assistant U.S. attorney. Even Yamaguchi acknowledged later that he had ``crawled into a shell.'' Veteran prosecutors were leaving the office in droves. Case filings and conviction rates were plunging. And criminal referrals from the SEC and FBI -- some involving serious allegations of fraud in the high-tech industry -- were going unprosecuted. Even the Cal Micro case, the office's best hope for sending Silicon Valley executives a message, seemed in jeopardy. A Day of Reckoning, 1997 Informix was at the threshold of disaster. On July 30, 1997, directors and top company executives gathered to hear the devastating news in the Palo Alto offices of Wilson Sonsini Goodrich & Rosati, the most powerful law firm in Silicon Valley. The board had already stripped White of his title as chief executive officer and hired a replacement, former 3Com executive Robert Finnochio, to review Informix's financial records. After asking White to excuse himself from the meeting, the directors listened raptly as Finnochio described the widespread accounting irregularities that White and other company executives had allegedly engaged in or condoned. Finnochio's grim report meant that White was finished. When Finnochio completed his presentation, the directors severed all ties with White and replaced him as chairman with Finnochio. It took Ernst & Young three months to determine the full extent of the questionable accounting schemes. On November 18, Finnochio made what he described as ``the mother of all financial announcements.'' Informix would have to restate financial results for the previous three years. The company disclosed that it had improperly claimed a staggering $278 million in revenues and $236 million in profits. On the brink of insolvency, Informix laid off thousands of employees. Offices were closed and operations consolidated. The company abandoned plans to build showrooms around the world

and announced that it would sell a 27-acre parcel of land in Santa Clara, the planned site for new headquarters. Immediately, Lerach and other plaintiffs' attorneys filed class-action suits on behalf of the thousands of investors who had lost millions of dollars, and Informix braced for the onslaught. THE EVE OF TRIAL, 1998 It was well past midnight in the warren of cramped offices on the 11th floor of the Phillip Burton Federal Building, and Crowe was still working. He had been at it for days, his frustration mounting. More than three years had passed since the investigation into Cal Micro had opened. In the middle of trial preparations, Leo Cunningham, then Richard Seeborg, the two prosecutors who had worked the case from the beginning, left the U.S. attorney's office to join big law firms in Silicon Valley. Crowe had replaced them, and now, in early 1998, he was handling the case alone. He looked over the 600,000 pages of documents with a mixture of disbelief and terror. ``I got started so late,'' he said. ``There was no secretary, no paralegal, no resources I could count on from the office -- nothing. ``At one point I had to decide between spending 80 hours with (a key witness) or going through every document. I chose the witness, but I was afraid the defense would find something in those papers, and I would be blindsided during the trial.'' Finally, the Department of Justice dispatched Pamela Merchant, a senior fraud specialist in Boston, to help him try the case. Meanwhile, fraud cases from the SEC and the FBI continued to pile up on his desk. But there was little Crowe could do. He tried to farm them out to other prosecutors, but he couldn't force anyone to take them. And he was acutely aware of how Silicon Valley would view the inaction. ``These big white-collar cases . . . should have high impact,'' he said. ``But if you screw around for four or five years, no one takes you seriously.'' An Empty Victory, 1998 The Cal Micro case finally went to trial in June.

Crowe and Merchant believed they had a strong case, a tour de force constructed around a stack of financial records and the testimony of two former Cal Micro executives who agreed to testify for the prosecution. In their defense, former CEO Desaigoudar and treasurer Henke told the jury that they had done nothing wrong, that they never saw memos or attended meetings where the fraud was discussed. And they repeatedly blamed the wrongdoing on their subordinates. Their testimony proved unpersuasive. On July 14, 1998, after a five-week trial, jurors found each man guilty of five felony charges, including conspiracy, securities fraud, insider trading and making false SEC filings. It was a major victory for the demoralized U.S. attorney's office. But for Crowe it was the end of the road. Just before the trial, his supervisor entered his office and complained that Crowe hadn't done anything with the mounting inventory of criminal fraud referrals. ``I exploded,'' Crowe said. ``I had given him the cases to reassign months earlier so I could concentrate on the trial, and he had done nothing.'' Crowe was sick with frustration. He had relished prosecuting Widergren, Desaigoudar, Henke and other white-collar criminals. He had even won an award for his representation of financial fraud victims. In August, he cleared out his desk in the federal building. When he submitted his resignation, he had no idea who would take over the inventory of cases the office still hadn't dealt with, including one potentially explosive case referred to him months before. The case against Informix. THE VERDICT, 1998 On December 8, a vast Pacific Gas and Electric Co. power failure left only dim, backup lighting in the San Francisco courtroom of U.S. District Judge Vaughn Walker. Despite the jury's verdict against Chan Desaigoudar and Steven Henke, the former Cal Micro executives refused to acknowledge their crimes, insisting that their subordinates had betrayed them. ``If I have a regret about my conduct as a former CEO and chairman,'' a defiant Desaigoudar told the judge, ``it is simply this: I might have trusted my employees. For this, I and my family will suffer and have been suffering.'' Walker tentatively ruled that he would impose prison terms of 36 months for Desaigoudar and 32 months for Henke, far less than the government had requested.

Merchant was furious. ``Your honor, the government is deeply troubled,'' she said. She reminded Walker that the fraud had cost investors tens of millions of dollars. ``This is a significant, as the court has said, grave matter.'' She urged Walker to reconsider and to send a strong message ``to the public and the investing community that this type of crime would not be treated differently than any other crime.'' Walker was unswayed. He praised the accomplishments of the two defendants. He extolled their civic and charitable contributions. He even called their situations tragic. ``This cannot by any stretch of the imagination,'' he said, ``be rendered equivalent to fraud which takes advantage of helpless and uninformed or particularly gullible individuals.'' EPILOGUE In August 1998, Yamaguchi resigned as U.S. attorney amid growing criticism from judges and attorneys about his ineffectiveness. Several months later he became an immigration law judge in San Francisco. Yamaguchi was immediately replaced by Robert Mueller, a tough, career prosecutor from Washington, D.C., who immediately promised to step up prosecutions of white-collar crime, focusing on Silicon Valley. A month later, after a four-year investigation, two former executives of Media Vision, a Fremont chip developer, were indicted on charges of securities fraud and insider trading. The case has yet to come to trial. Mueller said the agency has ``a number of cases in the hopper,'' but he declined to comment further. In July of this year, the shareholders' class-action suits against Informix were settled. Without admitting any wrongdoing, the company and Ernst & Young agreed to pay investors a total of $142 million, the largest securities fraud settlement in Silicon Valley history. White, who was not held personally liable for any of the settlement, currently lives in Atherton and sits on the boards of several companies, as well as his college alma mater. He denies any wrongdoing and has declined to comment further. Howard Graham, who also paid nothing under the settlement, refused to comment. No criminal charges have been filed against White, Graham or any other Informix executives by the U.S. attorney's office. Desaigoudar and Henke have appealed their convictions, and they remain free on bail.

Crowe practices law in San Francisco, representing investors. INFORMIX Roger Sippl founded Informix in 1980 after Hodgkin's disease forced him to put his medical education on hold. The company quickly grew into one of Silicon Valley's hottest companies, and after Phil White took over in 1989 it seemed poised to surpass Oracle, Sybase and other competitors as the top maker of data-base management software. A troubled fling with a multimedia-database product and charges of accounting fraud and illegal insider trading derailed the company in 1997. The company has attempted to recover ever since, and last year posted net income of $57.7 million after a $357 million loss the year before. CALIFORNIA MICRO DEVICES Based in Milpitas, California Micro Devices makes silicon chips, filters and electronic circuitry for workstations and personal computers. Since its former top officers were accused of securities fraud in 1994 and then convicted last year, the company has attempted to recover by expanding into new markets. For the fiscal year ended in March, Cal Micro lost $2.8 million on sales of $33.6 million. It has 150 employees in Milpitas and 110 in Tempe, Ariz. In June 1997, Shyam Das sat down for sodas with venture capitalist Jeffrey Drazan at the Peppermill Restaurant in Cupertino. Drazan and his firm, Sierra Ventures, had put about $7 million into the company Das founded the year before, and after a few minutes of small talk, Drazan said he wanted another Sierra director on the board. Das was wary. For years, he had worked alone to perfect the company's product -- an ingenious device capable of storing more data on a computer disk than anyone thought possible. And he could feel Drazan prying it away. But Drazan assured him they would ``always be together,'' so Das relented. It was a critical mistake. Two months later, Das was out, fired from his own company -- with Drazan's new director casting the deciding vote. And although Das still held millions of shares of company stock, he said the directors eventually found a way to betray him one more time. ``When I handed over this company,'' Das said, ``it was just like losing a child.''

Every year, hundreds of entrepreneurs like Das create high-tech companies with the hope of striking it rich. Although most know the risks of the open market, few expect to lose their startups to the professionals they depend on for advice and support. But over the past decade, some of the high-tech industry's most prominent venture capitalists and lawyers have been accused of stealing companies from the entrepreneurs they agreed to finance and of betraying clients they were hired to represent. ``So much of this is happening out there,'' said Santa Clara Superior Court Judge Conrad Rushing, who presided over a trial involving several venture capitalists. ``What I see in lawsuits is just the tip of the iceberg.'' Silicon Valley's venture firms control an estimated $22 billion of capital, and in the roiling, high-risk markets of the high-tech industry, they're not afraid to seize control of a company or force its founder out. But to call them to account would be financial suicide for almost any entrepreneur. ``Suing venture capitalists is like suing the Mafia,'' said Michael Kalashian, an entrepreneur who wrested a $15 million settlement from venture capitalists four years ago. ``In the valley here, anyone who is in business somehow interacts with the VCs . . . and they're terrified.'' The power of Silicon Valley's top venture capitalists is rivaled only by the elite attorneys who represent them, the partners at megafirms that also represent virtually all of high-tech's largest corporations. Inexperienced entrepreneurs hire these firms at their peril, often discovering that the attorneys have sold them out for more profitable clients. ``It's worked like that down here for years,'' said a San Jose attorney who defends large high-tech companies in lawsuits. ``Is it unethical? Well, to use a legal term: Hell, yes.'' This is the story of how some venture capitalists and lawyers operate in Silicon Valley and how they pay little heed to laws prohibiting conflicts of interest and the abuse of minority shareholders. It is also the story of some of the entrepreneurs and engineers who were ruined and left behind. UNDISCLOSED LOYALTIES, 1987 Narpat Bhandari had a brilliant idea for a lightning-fast computer chip. So he went to the most powerful law firm in Silicon Valley, believing that its contacts within the high-tech industry would help get his enterprise off the ground. But he says he soon learned what can happen when connections become conflicts -- and lawyers must choose between rival clients.

It all started in January 1987, when Bhandari, chief engineer of Fairchild Semiconductor's new products division, decided that after spending three decades as a chip designer, he would start his own company. Business plan in hand, he and a partner hired Wilson Sonsini Goodrich & Rosati, the influential Palo Alto law firm. Name partner Larry Sonsini helped Bhandari revise the business plan, counseled him on resigning from Fairchild and made sure the new chip did not violate any patents, Bhandari says. Over the next few months, Sonsini escorted Bhandari through the venture capital firms along Sand Hill Road in Menlo Park. Sonsini made a particular point of introducing Bhandari to John Doerr, Silicon Valley's premier venture capitalist and a director of high-tech giant Cypress Semiconductor. Doerr pitched Bhandari's idea to Cypress, and the company was intrigued. It hired Bhandari as a consultant, and after months of negotiations, says Bhandari, orally agreed to pump $13.6 million into his startup, pay him $2 million and grant him stock in both the new company and Cypress. Sonsini incorporated Bhandari's Aspen Semiconductor and, according to Bhandari, promised to document the Cypress deal. But in August, Bhandari says he made a disturbing discovery: Sonsini was not only Cypress' attorney and secretary, he had incorporated and guided the company through an initial public stock offering and was representing its chief executive officer, T.J. Rodgers. When Bhandari complained about the conflict of interest, Sonsini assured him that all was well. The attorney even suggested that his relationship with Cypress and Rodgers would help, according to court documents. Bhandari did not want to lose Sonsini's influence, so he says he continued to retain Sonsini as legal counsel, even appointing him to Aspen's board of directors. All Bhandari could do was hope for the best. -- -- -- A calm and dignified counselor to a new generation of brash tycoons, Sonsini has been called the ``consigliere of Silicon Valley.'' At 58, he is one of the most powerful figures in the high-tech industry. He is a confidant, a matchmaker and a financial and legal strategist. He has drafted business plans for some of Silicon Valley's best known companies and introduced the partners of some of its most prestigious venture capital firms.

Born in the farm country of upstate New York, Sonsini came west to California with his family in 1949. He graduated from Berkeley's Boalt Hall law school in 1966. But rather than join a lucrative New York practice or clerk for a federal judge, he took a job with an obscure law firm run by John Wilson in a no-name valley south of San Francisco. It was the best move Sonsini ever made. He saw the possibilities in that valley, the potent, untapped energy of young companies and the freedom of a culture not yet formed. ``It was the idea,'' he said, ``of being able to build something.'' At first, as the fourth attorney in a four-lawyer office, Sonsini handled as many divorces and slip-and-fall cases as incorporations for entrepreneurs. But in the early 1970s he represented a company called ESL, which was run by a future secretary of defense named William Perry. ESL wanted to sell its stock publicly and was about to hire a major San Francisco firm when Sonsini intervened. ``I said, `I can do this, and I can do it better,' '' he recalled. ``I just hated the idea of losing clients.'' ESL stayed with Sonsini. So did Seagate Technology and Intel and LSI Logic, and before long, over the qualms of some of his partners, Sonsini had built a law firm that could serve almost every need of companies in the rapidly emerging high-tech industry. Then, in 1980, Sonsini handled one of the best-known deals in high-tech history: The initial public offering of Apple Computer's stock. On December 12, Apple's shares rose by 30 percent, making dozens of Apple employees paper millionaires. Founder Steve Jobs' stock alone was worth $217 million. Suddenly it seemed as if every high-tech company and entrepreneur wanted to be Sonsini's client. ``Now they weren't calling us out of loyalty,'' he said. ``They were calling us because we were the best.'' It wasn't long before the firm was in such demand that it occasionally found itself representing parties on opposite sides of a deal, a conflict of interest that ethics rules expressly prohibit unless waived by clients. Sonsini acknowledges that ``the technology industry is unique'' in its potential for legal conflicts. But he says ``it really isn't an issue. We are rarely in the same deal with two clients.'' And if they are, he says they ask clients for waivers, which many seem happy to grant. Others, though, tell a different story -- of an untold number of entrepreneurs and engineers afraid to talk about what the law firm did to them.

``The number of conflicts cases (Wilson Sonsini) has settled so they never see the light of day is incredible,'' said one entrepreneur. ``There's considerable fear of Larry Sonsini in the valley, fear by entrepreneurs who don't want to get blackballed.'' -- -- -- It was Sonsini's reputation as the top attorney in Silicon Valley that prompted Bhandari to appoint him an Aspen director -- despite the lawyer's long-standing relationship with top Cypress officials. Six months later, Sonsini and the other Cypress directors fired Bhandari as chief executive officer of Aspen. And because Sonsini failed to document Bhandari's original deal with Cypress, Bhandari says he received none of the cash or stock he was promised. As a final indignity, Fairchild, Bhandari's former employer, sued him for trade-secret theft. Cypress CEO Rodgers said he fired Bhandari because five Aspen managers complained of a ``leadership problem'' and the company was three months behind in creating its project plan. Sonsini contends that he told Bhandari to get a new lawyer in April 1987, according to court documents. Sonsini said he stopped representing Bhandari in May. But Bhandari says he believed until his firing that Sonsini was his attorney. There was nothing in writing to indicate otherwise, he says. He also says he believes Cypress dumped him to gain complete control of the chip technology he had created and the engineers and management team he had assembled. ``The smaller you are,'' Bhandari said, ``the more they think you are nothing.'' Bhandari spent the next eight years fighting in the courts. He sued Cypress, Rodgers and Aspen -- all defended by Wilson Sonsini -- and in September 1991 won a $750,000 judgment. He defended himself against Fairchild, since acquired by National Semiconductor, and the company eventually dropped the case. Finally, he sued Wilson Sonsini for malpractice, and after years of rancorous motions, the suit was settled confidentially. A NEGLECTED DEFENSE, 1989 When Greg Galloway's thriving disk drive company was sued for trade-secret theft, he hired Wilson Sonsini to handle the case.

But according to his allegations in a lawsuit, Galloway would soon discover that retaining the valley's top law firm was no guarantee of an aggressive defense -- particularly if the firm's interests potentially conflicted with its client's. Galloway's company was called Tronix Peripherals. In 1989, its competitor, Trace Products, accused it of stealing software code by hiring the code's author, a former Trace employee. But almost from the beginning, the Wilson Sonsini attorneys made a series of puzzling mistakes, according to Galloway's lawsuit. They discouraged Galloway and Tronix from reaching a settlement that might have limited the damages. They delayed filing a summary-judgment motion that could have cut the case short. They essentially admitted Tronix's liability in a letter to one of the company's own customers. The alleged missteps worried Galloway, but he says that when he asked about them, he was told to concentrate on running his business. Then Galloway learned that Wilson Sonsini represented a company called Applied Biosystems. The author of the secret source code had worked at Applied Biosystems after leaving Trace and before joining Tronix. While at Applied Biosystems, and possibly at other firms, he had used the disputed source code, Galloway claims. And if the author of the source code had used it at Applied Biosystems and elsewhere, it was no longer secret when Tronix obtained it -- meaning Tronix could not be guilty of trade-secret theft. But the Wilson Sonsini lawyers failed to pursue that defense, Galloway claims, because they did not want to expose another client, Applied Biosystems, to legal liability. Galloway says Wilson Sonsini never asked him to waive the conflicts created by its representation of Applied Biosystems, or another company with ties to his suit. But the worst lay ahead. On the eve of trial, Michael Ladra, the lead attorney on Galloway's case, abruptly announced that he would no longer handle the lawsuit, claiming that another client needed his immediate attention, according to Galloway. Wilson Sonsini attorney Peter Courture took over, but he was ``totally unfamiliar with the facts, issues and evidence,'' Galloway claimed in court papers. The trial was a disaster. In March 1991, a jury in Santa Clara Superior Court found Galloway liable to Trace for $1,386,749 in damages. Galloway also owed Wilson Sonsini $135,627 in legal fees. The debts ruined Galloway. A year later he filed for bankruptcy, and Tronix closed its doors.

Galloway eventually sued Wilson Sonsini for malpractice, and the case was settled confidentially in 1996. Ladra says Galloway's claims were ``silly'' and contends the firm didn't pursue the Applied Biosystems angle for many reasons, the main one being that ``it would have actually been contrary to Galloway's interests.'' Ladra also says Courture was fully capable of trying the case and Galloway was urged to settle ``numerous times.'' ``The bottom line,'' he said, ``is the guy lost the case, and so he decided to blame someone other than himself.'' Galloway and Courture will not talk about the Tronix case or the malpractice suit. ``The one thing I have learned about legal battles,'' Galloway said, ``is they have nothing to do with truth and justice but only how you develop the facts.'' PILFERED SECRETS, 1991 It was every entrepreneur's worst nightmare. One day EP Technologies was seeking venture capital financing to expand its medical equipment business. The next day an entirely new company -- financed by the same venture capitalists -- arose to make exactly the same product for exactly the same market. EPT's nightmare began in 1990, when the company decided to expand its medical equipment business and look for financing. After a three-month search, it contacted Sierra Ventures. Founded in 1982, Sierra enjoyed a reputation as one of the most aggressive venture capital firms along Menlo Park's Sand Hill Road, headquarters for Silicon Valley's most prominent financiers. ``The ongoing joke is that they're greedy and avaricious -- and proud of it,'' said one former investor in a Sierra fund. Sierra was intrigued by EPT's products for diagnosing and treating heart disease. So it asked for more information and received a stack of confidential documents describing the company's business and technology. Over the next two months, Sierra studied every aspect of EPT. It hired one consultant to evaluate the company's technology and business potential and another to investigate the company's patents. It even introduced businessman Harry Robbins to EPT as a potential chief executive officer.

On Jan. 15, 1991, Sierra gave EPT a financing proposal. In March, the company responded with a counterproposal. Then the deal went dead. On April 10, Sierra abruptly told EPT that there would be no further negotiations. Company officials were stunned. But five days later they received even worse news: Sierra had agreed to finance CardioRhythm, a newly formed company that planned to compete directly with EPT. CardioRhythm was founded by Robbins and the two general partners of the financing agent that had introduced EPT to Sierra -- George Savage and Andrew Thompson of Savage-Thompson Management. In exchange for Sierra's backing, the three founders granted the firm stock and seats on the board. And it only got worse for EPT. When the company asked another venture capital firm for financing, Sierra tried to persuade the firm to invest in CardioRhythm instead, according to court documents. And later, when EPT discussed a merger with medical manufacturer Medtronic Inc., Sierra intervened and persuaded Medtronic to buy CardioRhythm. Finally, on March 25, 1993, EP Technologies sued Sierra, accusing it of using EPT's confidential information to help start CardioRhythm and then thwarting EPT's merger with Medtronic. Less than a month later, the lawsuit was settled confidentially. The lawyers for EPT declined to discuss the case. Savage, Thompson and Robbins could not be reached for comment. Sierra general partner Jeffrey Drazan said he did not recall the details of the case. In 1996, EP Technologies was acquired by Boston Scientific. SUDDEN INTERFERENCE, 1994 ASCNet was ready for the market. The company had developed a wireless phone system to connect motorists with a 24-hour help line and road service, and several organizations serving seniors and motorists were already promoting the product. But when word of the company's wireless technology reached Drazan at Sierra Ventures, ASCNet claims it quickly found out what can happen when venture capitalists want in on a deal.

Drazan had come to Silicon Valley from New York in 1984, a Long Island doctor's son with an engineering degree from Princeton and an MBA from New York University's Graduate School of Business Administration. At Sierra he prided himself on his ability to move fast, to decide quickly whether a company deserves funding and then to assert control. His approach was often successful, but some company executives refused to work with him. ``He is fiery tempered, quick to draw,'' said one company founder who turned down financing from Sierra when Drazan insisted that the firm be the sole investor. ``He just brazenly thinks that he's smarter than any of the entrepreneurs in his portfolio companies.'' Drazan thought ASCNet's ``Road Phone'' might be useful to one of Sierra's other investments, a Sunnyvale firm called Air Communications, which was struggling to develop a combination cellular phone, fax and modem. Over the next few months, Drazan and Air Communications chief John Soden talked to ASCNet about licensing the wireless technology. They even raised the possibility of acquiring ASCNet. But the Road Phone was ready for sale and ASCNet was running low on cash. On August 30, 1994, it agreed to accept $1.5 million from Nassgil Financial. The next day, Drazan and Soden called Nassgil and asked for a meeting, according to court documents. In the afternoon, they sat down with Nassgil's president, Gabriel Nassar, and its attorney, Paul Konalpelsky, and asked whether Sierra Ventures and Air Communications could get a piece of the ASCNet deal. When Nassgil said no, Drazan allegedly threatened to block its investment in ASCNet. He told Nassar and Konalpelsky that ASCNet's board was in disarray and that he commanded the support of the company's shareholders and directors. Unnerved by the threat, Nassgil withdrew its offer of financing. Within a month, ASCNet was out of business. Drazan disputes the allegations, which are contained in a lawsuit filed against Sierra Ventures and Air Communications. He said he met with Nassgil at ASCNet's request, and after a one-hour discussion he left with the possibility of a Sierra investment still open. ``I went to Australia, and when I got back there was a lawsuit on my desk,'' he said. ``It was ridiculous.'' The lawsuit accused Drazan and Sierra Ventures of illegally interfering with ASCNet's deal with Nassgil. It was settled confidentially, and Konalpelsky refuses to comment on the case.

Later, Sierra unloaded its investment in Air Communications, deciding, says Drazan, that the company's product ``was a silly idea.'' SHIFTING ALLEGIANCES, 1995 Donald Sollars knew he had a problem. But he never expected to be dumped abruptly by his attorney, who Sollars claims sided with the three former partners trying to gain control of his company. Lawyers say it happens all the time in Silicon Valley: An entrepreneur trusts the attorney he hired to continue to represent him, only to find that the attorney has chosen instead to represent a company, its investors -- or even the people the entrepreneur has engaged in battle. Sollars had worked at Intel and Sun Microsystems designing computer chips. In his spare time he had invented a unique component for microprocessors used in cell phones. He believed the device was good enough to be the foundation of a profitable business, but he also knew he needed help starting a company. ``I was a techie,'' he said, ``with zero business experience.'' So in March 1995 Sollars and his colleague Krishna Yarlagadda turned to Wilson Sonsini, hiring John Goodrich at the firm to help them and two other partners start a company called Infinite Solutions Inc. Goodrich incorporated the new company in April, but by then Sollars and the other three partners had started quarreling. They couldn't decide how to divide control and ownership of the company, how to share its assets or how to structure its business. The disagreements eventually made it impossible for Sollars to proceed, so in December 1995, he resigned from Infinite Solutions. And because Infinite Solutions was based on his technology, he demanded that the company be dissolved. But according to allegations in court documents, Yarlagadda refused to give up the technology. In fact, he told Sollars, he and the other two partners intended to use it in a new business with fresh investors. Sollars and Yarlagadda took their disagreement to Goodrich, who informed them that Sollars didn't own the disputed technology and should negotiate a settlement among the partners, the court documents say. But when negotiations started, Sollars says he discovered that Goodrich was representing Yarlagadda and the other partners against him. Sollars was furious at what he viewed as his lawyer's betrayal. He hired another attorney who demanded that Wilson Sonsini stop representing the three partners.

But Goodrich and Wilson Sonsini refused, according to the documents. Goodrich denied that he had ever represented Sollars individually. The lawyer claimed that he represented Infinite Solutions -- even though he had provided legal advice to Sollars and the other partners long before the corporation existed. Finally, Sollars sued to stop Wilson Sonsini from representing his former partners -- and after months of negotiations, the law firm withdrew. Goodrich and Sollars refuse to comment on the case, and Yarlagadda could not be reached for comment. ``Let me be blunt,'' Sollars said, ``the last thing I want to do is to piss off John Goodrich and Wilson Sonsini.'' SHAREHOLDERS AGREEMENT, 1998 Vijay Parekh figured he had an ace in the hole, a shareholders agreement designed to preserve his control of the software company he founded the year before. But when the venture capitalists decided they wanted him to step down as CEO, Parekh discovered how little the document meant. His company, Manage.com, made software for monitoring computer networks. It had already received a multimillion dollar investment from U.S. Venture Partners when, in July 1998, the firm decided that Manage.com needed more money -- and a new CEO. Lucio Lanza, U.S. Venture's partner in charge of the company, told Parekh not to worry, according to court documents. He said Parekh would continue to sit on the company's three-man board of directors, and he promised Parekh the job of chief technology officer. It would all be taken care of at the July 13 board meeting, Lanza said. But at the meeting, Lanza moved to add a fourth director -- his partner Jason Green. It caught Parekh completely off guard. Appointing a new director wasn't on the meeting's agenda. So Parekh protested, stressing that the company's shareholders agreement specifically prohibited the election of an additional director without the unanimous consent of the original three. The board turned to Manage.com's lawyer, Thomas DeFilipps of Wilson Sonsini. To Parekh's astonishment, DeFilipps told the board that it could appoint Green without Parekh's consent, the court documents say. A vote was called. Parekh says he voted against Green's appointment, but the other directors voted in favor.

Several weeks later, Parekh received a copy of the board minutes drafted by DeFilipps. He was stunned to see in a section titled ``Election of a New Director'' that his vote against Green's appointment had been recorded as an abstention. Parekh dashed off a letter to DeFilipps. It demanded that the minutes reflect Parekh's vote against Green and that the attorney explain in writing why Green's appointment didn't violate the shareholders agreement. Three days later, just hours before Manage.com's next board meeting, Lanza stepped into Parekh's office and fired him. Parekh was crushed. He had lost not only his job and his company, but about 600,000 shares of Manage.com stock scheduled to vest in six more months. On November 10, Parekh sued U.S. Ventures, Lanza and Wilson Sonsini for allegedly violating the shareholders agreement and firing him without cause. Although Parekh says the case is being settled, he refuses to discuss it further. Lanza and DeFilipps also decline to comment on the case. A PAPER FORTUNE, 1999 First they gave Shyam Das millions, then they stripped him of everything. After 12 years at Digital Equipment Corp. and a few months at California Micro Devices, Das decided to start his own company. The 51-year-old engineer had spent years working alone on the prototype of a new data storage device. And in early 1997, he began looking for financing to construct a new manufacturing facility. In February, Drazan and Sierra Ventures agreed to invest about $5 million in Das Devices, and Das figured he was in business. He was also in trouble. As soon as Drazan joined the company's board, Das' role began to change. His title fell from president to executive vice president. He lost much of his responsibility over the company's technology. The influx of venture capital and other investments diluted his share of the company's stock from 50 percent to 5 percent. In June, Drazan allegedly told Das that he wanted to appoint a Sierra investor named Bert Zaccaria to the board. Das reluctantly agreed, but only after Drazan assured him that ``you, me and Bert will always be together.''

Drazan says he does not recall the conversation, but wouldn't say such a thing because ``every director is independent.'' A few months later, as the August 1 board meeting was about to begin, the company's CEO handed Das a letter, informing him that he had just been fired. When Das protested, the directors voted 3 to 2 for his ouster -- with Zaccaria casting the deciding vote. Drazan contends that Das left the company ``by mutual agreement'' and received ``a very generous exit package.'' Das says it was small compensation for almost five years of work and the loss of his dream. His only consolation was that he still held more than 6 million shares of stock in the company, potentially worth a small fortune if the company was ever sold. It wasn't long before a deal was in the works. In 1998, Applied Magnetics Corp. agreed to buy Das Devices, and Drazan and the other investors started calculating how to divide the proceeds. With the help of company counsel and Wilson Sonsini partner Adam Dolinko, they negotiated an agreement that called for loans to be paid first, then the noncompete agreements, then the series B through E preferred stock and, finally, the series A preferred and the common stock. The deal went down on Feb. 11, 1999, with Applied Magnetics paying about $65 million in stock for Das Devices. But after trickling through the descending levels of debt and equity, there was just enough money for the board of directors, Sierra Ventures and the other investors. Das got nothing. His more than 1.7 million shares of series A preferred stock and 5 million shares of common stock were worthless. ``I was so disgusted, that I decided never to work in that industry again,'' said Das, who has started a new company, eHoHo.com, specializing in selling houses and household products. Dolinko refuses to comment on the transaction. Drazan acknowledges that it ``wasn't perfect.'' But he says it ``turned out to be a bad deal for everyone. ``We lost our shirt on this thing, and I think it's somewhat pathetic that this guy doesn't appreciate that we did everything we could to save his company.'' THE ART OF THE DEAL

It is difficult to believe that anything like a bad deal could ever cross the desk of Larry Sonsini. The polished top is piled with sets of legal documents as neat as decks of playing cards. The rich green carpet beneath is spotless. The surrounding leather chairs and Chinese paintings are modest and perfectly tasteful. And yet Sonsini apologizes for the mess. Today Wilson Sonsini Goodrich & Rosati represents more than half of Silicon Valley's 150 largest public companies. The firm has more than 600 lawyers and expects to hire about 60 next year. A few firms are bigger, ``but none that can throw this kind of firepower at a business of any size,'' Sonsini said. ``I work harder now than ever,'' he said. And making millions of dollars a year has its advantages. Like the home in Woodside, the Porsche and Mercedes, the good cigars and the ``little trips away.'' But nothing surpasses the pleasure of being the most sought after lawyer in Silicon Valley. ``If you are passionate about what you do,'' Sonsini said, ``you are always trying to perfect your art.'' SIERRA VENTURES Founded in 1982, Sierra Ventures is among the largest venture capital firms in Silicon Valley, managing more than $500 million in funds. Sierra is run by four general partners and attracts investors -- called limited partners -- ranging from university endowments to prominent executives in the technology industry. The firm concentrates on investing in health-care and information-technology startups. One of its most successful deals was the 1986 financing of StrataCom Inc., a San Jose telecommunications-equipment maker that was sold to Cisco Systems in 1996 for about $4.5 billion. WILSON SONSINI GOODRICH & ROSATI Based in Palo Alto, Wilson Sonsini Goodrich & Rosati represents more than half of Silicon Valley's 150 largest public companies and last year earned profits of $620,000 per partner. Its

longtime clients include some of the best-known names in the technology industry, from Seagate to Apple to Hewlett-Packard. The law firm announced yesterday that it had elected 23 new partners, bringing the total number of partners to 140 out of more than 600 attorneys. Since 1978, the firm has invested in hundreds of startups that it has taken public. It is a technology so small that it cannot be seen with the naked eye and yet one so powerful that it could transform the data storage industry. It is a microscopic funhouse of mirrors and silicon crystals that can bounce a laser beam off a disk and, eventually, send information to and from a computer. It is ingeniously precise - and extremely valuable. And when inventor Ronald Maynard showed it to five entrepreneurs gathered in a Milpitas conference room April 2, 1996, a decade of arduous research seemed ready to pay off. But after months of negotiating a deal for the technology, Maynard says the entrepreneurs abruptly told him it belonged to them. When he protested, their lawyer threatened legal action, and all negotiations broke off. One year later, Maynard's deepest fears came true: The entrepreneurs had based their company on his technology - and then sold it for $325 million. In cash. Maynard got nothing. "I spilled my guts to these guys for many months," Maynard said, "and I came away with, 'You back off, or we'll sue you.' " It is the kind of tale told often in Silicon Valley. Technological breakthroughs, revolutionary concepts, visions of the future - intellectual property is the coin of the realm in the technology industry. And high-tech companies are willing to do almost anything to get it, no matter the boundaries of law or ethics. Some of Silicon Valley's most successful companies and executives have been accused of breaching confidentiality agreements, stealing secret documents, manuals and source codes, and taking proprietary software by hiring away the engineers who created it for competitors. Relatively few cases, though, have ever made it to criminal court, because they are expensive and difficult to prove. "We probably prosecute more of these cases than anyone," said Santa Clara Deputy District Attorney Julius Finkelstein, "and we're extremely selective."

This is the story of three classic cases of alleged trade secret theft in the high-tech industry: a billion-dollar empire built on disputed software; a unique Internet search engine pirated away engineer by engineer; and a technological advance potentially so valuable that one of the most prominent executives in Silicon Valley succumbed to its allure. The story begins in an office park outside Pleasanton, with a man who signs his e-mails, "DAD." PEOPLESOFT: THE BILLION-DOLLAR EMPIRE Get past the white hair, the comfortable grin, and the peppy voice and attitude, and there's still the love for stray cats and dogs and the five adopted children at home. Few would dispute that David A. Duffield is a caring man. And PeopleSoft, the Pleasanton company he founded in 1987 and ran until going part time a few months ago, reflects his relentlessly sensitive style in its culture, which employees typically compare to a family. But the origins of the company were anything but mom and pop - and Duffield is far from a pushover. PeopleSoft was born of a protracted battle over software - computer programs that Duffield allegedly stole from his previous startup and made the foundation of a new company he eventually guided to more than $1.3 billion in annual revenue. PeopleSoft and Duffield "have had a long record of success," said a former executive of Integral Systems Inc., the company Duffield left behind. "But we believe they did not have an honorable beginning." Growing up in Ho-Ho-Kus, N.J., Duffield got his first taste of sales by selling Christmas cards and gladioli door-to-door. He also excelled in math, and after entering Cornell University, he "fell in love with programming" computers, he says. In 1963, he earned electrical engineering and business degrees from Cornell and began selling computers for International Business Machines. He was content to spend his career at IBM until a co-worker hooked him on developing a computer program for scheduling final exams at universities. Duffield spent his weekends selling the program, and demand was so strong that in 1968 he and his colleague left IBM to form their own company, Information Associates. In 1972 the partners split, and Duffield founded Integral Systems. At first, Integral made software to help colleges manage employee benefits and payroll. Then, in 1979, it began developing similar programs for corporations. In 1984, Duffield helped adapt the programs to additional operating systems and databases. And by 1986, its annual sales topping $55 million, the company was producing software to manage entire personnel departments.

But Duffield saw a problem: The programs ran on mainframe computers rather than on the increasingly popular personal computers. He wanted to change the software once again, but his colleagues balked. "They said the mainframe business was going well," Duffield recalled, "and they didn't want to invest any more." So he and his wife, also an Integral employee, mortgaged their house in Blackhawk and raised $2 million in seed money. They then took the chief technology officer, a code writer, a technical writer and four other employees from Integral and, in April 1987, launched PeopleSoft. STRIKING SIMILARITIES, 1989 The parting, at first, was amicable. Duffield agreed in writing that Integral owned the software he had worked on at the company and that he would never disclose Integral's trade secrets or other protected information. The Integral workers who followed him to PeopleSoft also signed confidentiality agreements. But in April 1989, when PeopleSoft's personnel-management program appeared on the market, Integral thought it looked awfully familiar. The organization and flow of information within the programs were virtually identical, Integral claimed. The section that processed payroll - producing checks and tracking taxes and other deductions - was similar to Integral's in at least 25 specific ways, said Integral. Similarities also existed between the two programs' human-resource sections, which arranged information about employees. Integral, for example, had arbitrarily combined workers' "skills and languages" on a single screen. PeopleSoft had done the same thing. Finally, said Integral, the programs' security systems - the mechanisms that limited access to certain individuals - were not only identical but shared features found in no other program. Integral executives were furious. The company had spent almost 20 years and millions of dollars developing its product, which cost about $500,000 to license and about $1 million to install. In less than two years, PeopleSoft had created a remarkably similar product - without documents to show where it had come from. On Sept. 10, 1990, Integral sued PeopleSoft for copyright infringement and trade-secret theft.

SECRET DOCUMENTS, 1990 Integral's case was simple: PeopleSoft had used Integral's software as a road map to create a product with the same look and feel. And though the program's code, the language telling a computer what to do, was different, the directions it gave to the computer were essentially the same. Duffield and his colleagues had helped create Integral's program, and witnesses claimed to have seen manuals, tables and other secret Integral documents in PeopleSoft's offices. Former Integral and PeopleSoft employee Carolyn Carpeneti, for example, testified that she had helped create PeopleSoft's human-resource software by copying a confidential Integral manual she got from Duffield's wife, Cheryl. She acknowledged that she had known the manual was secret because Cheryl had told her "to be sure to lock it up in my desk when I go home at night. . . . She said nobody should see me using it." Cheryl Duffield has denied giving Carpeneti a copy of the Integral manual. But Integral's evidence would not persuade U.S. District Judge D. Lowell Jensen in Oakland to issue a preliminary order blocking PeopleSoft from selling its software. On July 19, 1991, Jensen ruled that Duffield and his colleagues at PeopleSoft clearly had access to Integral's software and supporting documents. But Integral had failed to prove the two programs were alike. Integral's biggest mistake: Hiring management consulting firm KPMG Peat Marwick as a technical expert. Experts were crucial in such a complex case, the judge said, and Peat Marwick's performance was "seriously flawed." For example, the firm had compared only similarities between the companies' programs, not shown significant differences, and then had tried to conceal its mistake. Jensen also said Integral was sloppy in protecting its secrets, having shown prospective customers how its software worked. And some of the company's purportedly secret information had already leaked out to the industry. So the judge refused to issue an order that, in effect, would have put PeopleSoft out of business. Integral pursued the case for almost five more months. On Dec. 13, 1991, PeopleSoft agreed to settle for approximately $1 million - with no admission of liability. KILLING THE ENEMY, 1992 For PeopleSoft, the amount would soon seem trivial.

The company's timing had been perfect: As it was introducing its new software, the market for personal computers and so-called client-server systems - which connect PCs to a central computer - was exploding. In November 1992, the company sold stock to the public and began minting stock-option millionaires. By 1994, revenues were $113 million. Four years later they were up to $1.3 billion. Last year, Duffield's stock alone was worth about $2 billion. He had so much money that his family foundation gave more than $200 million to end the killing of unwanted dogs and cats. And even after an industry slowdown late last year sent PeopleSoft's shares plummeting, the company, with a workforce almost cultlike in its loyalty, was the envy of corporations throughout the nation. Integral, meanwhile, never caught up. Its plans to go public fell through. It didn't release a personnel-management program for personal computer networks until 1993. The next year, CSA Holdings Ltd. in Singapore bought a majority of the company and today owns 80 percent of the stock. Although Integral touts itself as an industry leader, the company has only 96 employees, and sales of just $45 million last year make it a distant rival to PeopleSoft. "We became a little too successful," said Duffield in explaining why Integral sued PeopleSoft. "I think the (Integral) CEO was embarrassed because the sales force complained that customers wanted this fancy new (PeopleSoft) technology more than the mainframe stuff." But there is a saying around PeopleSoft, a sort of motto, according to Duffield. It goes, "Have fun, kill the enemy." NETSCAPE: THE PURSUIT OF GURU Michael Kahn's childhood dream was to be Albert Einstein. The fantasy would inspire him to design nuclear reactors and, eventually, sophisticated high-tech devices. It would not, however, prepare him for the cutthroat tactics that stripped his company of engineers and trade secrets. Kahn grew up in Springfield, Mass., and studied nuclear physics at Rensselaer Polytechnic Institute and Carnegie Melon University. He first worked at Westinghouse creating nuclear reactors for the U.S. Navy, a job he would have enjoyed but for the hours. "The odd side effect of working on secret projects was you had to leave at 5 p.m.," he said. "I left the company so I could work harder." He attended Stanford University business school, graduating in 1979, and worked for several Silicon Valley companies with products ranging from image processors to nuclear medical

equipment to an artificial-intelligence system for drafting legal documents. At most of the companies, he was an executive rather than an engineer because, he said, "In Silicon Valley, the truth is that engineers are treated like fodder." In 1992, Kahn turned full time to MicroGuild Inc., his own consulting firm. His plan was to use the profits from MicroGuild to develop a database search engine, software that would soon drive the expansion of the Internet. Kahn hired a team of top-notch engineers, and within three years he had created Guru. Unlike other search engines, Guru could tell the difference among legal, medical and other types of documents and thus avoid wasting a user's time with irrelevant information. It did not take long for word of Guru's unique features to spread through the valley, and in July 1995, Kahn got a call from Netscape. SEDUCING THE ENGINEERS, 1995 Netscape had been in business for about a year and had just released its first product: Netscape Navigator, software for browsing Web pages. But the company was dissatisfied with its Internet search engine and wanted to hear more about Guru, according to court documents. During a meeting with Kahn and his chief engineer, Rick Henderson, Netscape's Mike Holtz and Mike Barbarino floated the idea of buying MicroGuild. Barbarino, who had worked with Henderson before, encouraged him and MicroGuild to consult for Netscape on a few projects. Kahn and Henderson were intrigued. They had heard that Netscape planned to grow by acquisition, and selling Guru to a big Internet firm seemed to make sense financially. Little did they know how big Netscape would become. On August 9, the company sold 5 million shares of stock in one of the high-tech industry's most spectacular initial public offerings. The company that had never turned a profit was suddenly worth $2.3 billion, and Kahn and Henderson were eager to make a deal. In January 1996, MicroGuild sent Henderson to help Netscape on a project. The results were so impressive that Netscape hired MicroGuild again in March. Meanwhile, the companies continued to discuss an acquisition. Netscape Vice President Ira Scharfglass allegedly told Kahn that the company usually bought startups just to get their top engineers - but MicroGuild could expect a premium price because it also had great technology. In June, negotiations heated up. With an increasing unhappiness over its search engine, Netscape Vice President Bill Turpin allegedly asked what MicroGuild was worth. MicroGuild responded with an estimate "in the low tens of millions," and Turpin responded by ordering his technical and marketing managers to start testing Guru.

Over the next several months, after signing a confidentiality agreement, the companies modified Guru to run on Netscape's system. MicroGuild dropped all other work, and in August the sacrifice seemed worthwhile: Netscape declared that in many ways Guru was superior to its current search engine, according to court documents. Then, without warning, negotiations stalled. And in September, Netscape hired away two MicroGuild engineers. COOPERATION COLLAPSES, 1997 Kahn was worried. He asked Turpin if the deal was dead, and, according to court documents, Turpin assured him that it was still alive, even requesting specific terms for a sale. About the same time, though, Netscape hired a third engineer from MicroGuild. On November 13, Kahn protested the hiring, and Turpin allegedly promised not to take any more engineers. Kahn then proposed that Netscape buy MicroGuild for $12.5 million, and, according to court documents, Turpin said the best time for an acquisition would be early 1997. The two companies continued to work on projects together, but in June 1997 all pretense of cooperation collapsed. That month, a fourth engineer left for Netscape. Another followed in July and one more in September. By fall, MicroGuild had no engineers left - and no way to stay in business. Worse, it had given Netscape the source code for Guru and was frantic to get it back. But Netscape, with quarterly revenues of more than $150 million, was no longer interested in negotiations of any kind. On May 15, 1998, MicroGuild sued, charging Netscape with violating the confidentiality agreement, stealing trade secrets, interfering with employee contracts and committing fraud. The case is scheduled for trial next year. Netscape has declined to comment. Guru, meanwhile, is in limbo. Netscape says it never used Kahn's search engine. Kahn says he has tried to sell it, but "having lost our key developers to Netscape, it's pretty tough to sell just a piece of code." THE NEXT DISK DRIVE When Ron Maynard was young, the workshop behind the family home in Tampa, Fla., wasn't just a place to pound nails and build toy boats with three brothers and a sister. It was a laboratory, a proving ground for the skills and inspiration that 25 years later would attract big money in Silicon Valley.

Inventing was in Maynard's bones. Maynard's mother dabbled with inventions and passed down to him the family recipe for gunpowder. Her father invented products for the American Thermos Bottle Co. Her grandfather invented artificial diamonds. As a child, Maynard had to invent his own toys. After settling in Tampa, he built rockets and telescopes with home-ground lenses. "I used to look at the moon," he said, "and think, gee, that's where I want to go." In 1978, his passion for space drove him to the University of California at Berkeley and a major in space science. He alternated school with work at NASA in Houston, and while there he filed his first patent, a design for a fluid-based gyroscope. After graduation and two more years at NASA, Maynard received a post-graduate scholarship to Stanford University in 1985. He studied micromachining, the process of creating microscopic devices through chemical and electronic means. Over the next seven years he would become one of the few experts in the field, working on a doctorate and a thesis about his controversial data-storage device. Then, in 1992, he encountered the harsh world of high-tech finance. During his research on micromachining, Maynard had developed an electronic wrapper for endoscopes: thin, lighted tubes used to examine internal organs. The wrapper contained electronic muscles that received electronic signals for steering the endoscope. The technology drew raves, and a prominent venture capital firm in Menlo Park begged for a detailed demonstration. After Maynard showed the firm's partners how the device worked, they offered funding for a startup to produce it. Maynard was ecstatic - until he read the fine print. The deal called for the venture capital firm to get 6 million shares in the new company, while he would get a paltry 5,000, Maynard said. Worse, if Maynard didn't proceed with the deal, the venture capital firm threatened to take the technology anyway. "It was either, you play ball, or you get crushed under the wheels," Maynard says. So Maynard gave his technology patents to Stanford and dared the venture capital firm to try to get them. The firm retreated, and Maynard would eventually get them back. BROKEN PROMISES, 1996

A few years later, Maynard started a new company - Microdomain - to develop micromachined devices. And in March 1996, he got a call from a Stanford University researcher named Jeffrey Wilde. Wilde and three other entrepreneurs had founded Quinta, the latest in a long line of companies trying to create the next-generation disk drive, the vital data storage unit in most computers. In April, Maynard squeezed into a car with Wilde and Quinta co-founders Joseph Davis and Steven Kitrosser, and they headed for the Milpitas offices of Read-Rite Corp., a maker of magnetic disk-drive heads and an investor in Quinta. From the moment Wilde had called, Maynard had stressed that his techniques and designs were absolutely confidential. And in the car, Maynard harped about secrecy, making everyone promise to abide by the terms of a mutual nondisclosure agreement he had with him. So Davis signed the agreement on behalf of Quinta, and Maynard signed on behalf of Microdomain. But Maynard didn't stop there. The law firm of Wilson Sonsini Goodrich & Rosati had been giving him patent and contract advice for years, and he was well-prepared. So when the four met Read-Rite officers Steve Stone and Peter Bischoff in the Milpitas conference room, Maynard extracted their promises to keep whatever he showed them confidential. Only then did Maynard sit down to talk. During the next few hours, Maynard and the others discussed what they could do for one another. Quinta's goal was to make disk drive heads so small and precise they could read even the most densely packed data from a disk. And it wanted Maynard's help. Maynard says he showed Quinta his micromirrors and a high-level head design and diagrammed his technology on a large white board. About a dozen more meetings followed. But on June 18, without warning, Davis allegedly told Maynard that the technology was a co-invention. Maynard was furious. He had spent much of his post-graduate education developing and refining the technology and was finally writing his doctoral thesis to describe it. To lose it now - to forfeit "all my early years," he said - would be unthinkable.

Maynard strongly protested Davis' claim, and after several weeks of bickering, the parties met July 16 to try and patch things up. But when Maynard realized that Quinta's patent lawyer was at the meeting, he walked out, shocked at what he contends was another attempt to steal his designs. A few days later, Maynard says he called Petri Vainio, a partner at a venture capital firm called Sierra Ventures who had heard Maynard pitch the endoscope wrapper. He confided his suspicions, and Vainio promised to speak with his partner, Jeffrey Drazan, who was overseeing Sierra Venture's investment in Quinta. But when Maynard called later, he says Vainio refused to interfere with a partner's deal and advised Maynard to "get a good lawyer." Negotiations broke off, and on Oct. 14, 1998, Maynard sued Quinta, its founders and Sierra for trade-secret theft. Quinta disputes Maynard's version of the facts in a cross-complaint the company filed in March, saying Maynard never disclosed any useful information and that his head design was "unworkable." The company contends Maynard stole and patented its design for a disk-drive head and says it offered Maynard a job but that he refused. Drazan, whose name appears on a patent involving the disputed technology, dismisses the meetings between Maynard and Quinta founders as a prolonged job interview. "He interviewed for a position at Quinta," said Drazan, "and if he had some proprietary idea, a job interview was not the place to show it." SHUGART'S GAMBLE, 1997 Quinta soon began to boast that its breakthrough disk drive technology could transform the data storage industry. In only a year of business, the company raised more than $50 million and attracted corporate investors such as Seagate Technology. During the past two decades, Seagate founder and CEO Alan Shugart had built the company into the world's largest disk drive manufacturer by acquiring state-of-the-art technology. And the more he heard about the potential of Quinta's device, the more he wanted it. On July 2, Seagate announced that it had agreed to buy Quinta for $325 million in cash and deferred payments. "Getting into that technology was important, absolutely," Shugart said, "and others agreed with me." In a press release, Seagate declared that the acquisition of Quinta would help it maintain its edge over competitors in the rapidly advancing disk drive industry. But the deal was a huge gamble.

For one thing, Quinta didn't have a product that was even close to being ready for the market. Without Maynard, the company hadn't been able to make the technology work, said Read-Rite's Bischoff. "We ran into some technical difficulties that were tougher than anyone thought," he said. "The technology kind of went nowhere. It was a disappointment." At the same time, the patent applications Quinta had filed included the technology that Maynard contended was his. Seagate either didn't know or didn't care. But Seagate's law firm - Wilson Sonsini Goodrich & Rosati - was on notice of Maynard's claims, because the firm also represented Quinta and was Maynard's regular patent counsel. At one point, Wilson Sonsini even asked Maynard to waive the firm's conflict of interest in representing both him and Quinta, Maynard said, but after he refused, the firm concluded that no conflict existed. Industry reaction to Seagate's acquisition of Quinta ranged from bafflement to pity. The head of a rival disk-drive manufacturer said the Quinta investors "sold Al a pig in a poke. They conned him." Even some of Quinta's investors were surprised that Seagate paid so much. "I think the Quinta guys were just very good salesmen," said Bischoff. "They snowed him." Quinta's founders and investors made millions on the deal. Drazan says it was one of his most successful deals at Sierra, emphasizing how he and the others held out for cash, even though Seagate was offering a higher price in stock. "Seagate stock was up in the 40s, and all the investment banks had buys on it," he said. "We were less optimistic." And for good reason. Several months later, Seagate's stock plummeted below $20 a share. "We have some people here," Drazan said, "who are pretty savvy about the disk drive business." A LEGEND'S DEMISE, 1998 It was the type of mission that Larry Sonsini had grown comfortable with.

For 30 years, he had served as legal confidant to elite corporate boards, trusted with delivering bad news when the occasion called for it. But this time, it was different. This time the focus was Shugart, one of Sonsini's first clients and closest friends. Seagate had been in a financial free fall for months. A week after the Quinta deal was signed, Seagate announced a 41 percent drop in fourth-quarter earnings from the year before. The next four quarters only got worse, with the company suffering a $530 million loss for the fiscal year. Shugart had already relinquished the title of Seagate president to former investment banker Stephen Luczo – amid assurances that the grand old man of the disk drive industry wasn't close to retirement. But then came the closing of the plant in Clonmel, Ireland, the 10,000 layoffs and the 10th position on the Wall Street Journal's annual list of the 25 worst-performing stocks. The directors of Seagate had seen enough. So they turned to Sonsini, the company's outside counsel and perhaps the only one who could make Shugart understand what had to be done. On Saturday, July 18, Sonsini phoned Shugart to ask if they could meet at Shugart's Carmel office. "We've always been very honest with each other, very candid," Sonsini said. "And if I delivered the message, it would be clear that this was not a negotiation, that this was a decision." Sonsini arrived in Carmel and greeted Shugart warmly. Then he delivered the news: Shugart had been fired from the company he had founded two decades ago. The company would give him a $10 million severance package, but he would have to step down as CEO and chairman of the board. Shugart smiles about it now, saying Sonsini "was trying to convince me that the deal was so good, I shouldn't make waves." But Shugart, who had resisted pleas to resign for many weeks, took the news hard. The next day, when Seagate's board met to make the firing official, the directors phoned him on a conference call. He had just one question: "I asked them why," he said. "And the only thing they said was, 'It's time for a change.' " QUINTA CORP. Founded in 1996, Quinta Corp. attempted to create a next-generation device for reading and writing information stored on computer disks, but it never developed a product.

Seagate Technology bought the company in 1997 for $325 million and made it a wholly owned subsidiary. One year later, Quinta was absorbed into Seagate's research and advanced concepts division. Today, Quinta's technology is still not in use and, according to a Seagate spokesman, won't be anytime soon. PEOPLESOFT Founded in 1987 as a maker of personnel-management software, PeopleSoft has grown into a dominant supplier of software for managing finances, human resources, manufacturing, distribution and other corporate matters. The company has more than 6,000 employees and in 1998 earned $1.3 billion in revenues. The past year has been tough for PeopleSoft, which has suffered a major drop in its stock and substantial layoffs for the first time. The downturn is industrywide but particularly unnerving for a company that has a house band and refers to its employees as PeoplePeople. Many in the audience were expecting a routine dedication speech. But shortly after Securities and Exchange Commission Chairman Arthur Levitt began his remarks, a hush fell over the executives and academics gathered at New York University's new Center for Law and Business. ``Too many corporate managers, auditors and analysts are participants in a game of nods and winks,'' said Levitt, the former head of the American Stock Exchange. ``Managing may be giving way to manipulation. Integrity may be losing out to illusion.'' Referring to the role of unreliable accounting in the recent Asian financial crisis, Levitt made clear what is at stake for the U.S. economy. ``Today, American markets enjoy the confidence of the world,'' he said. ``How many half-truths, and how many accounting sleights of hand, will it take to tarnish that faith?'' The chairman's ``Numbers Game'' speech in September 1998 was the opening volley in the SEC's vaunted campaign against improper and sometimes fraudulent financial reporting in corporate America. And dozens of the most prominent companies in Silicon Valley were among the agency's targets. But Levitt and the SEC would soon discover just how difficult it is to crack down on the financial abuses thriving in Silicon Valley -- and just how powerful high-tech executives have become.

Since the days of the valley's first shoestring startups, the industry has emerged as one of the most potent political forces in the nation's capital. Over the past five years, Silicon Valley executives have waged a relentless campaign to weaken the laws that regulate them. In 1995, the executives helped persuade Congress to pass the Reform Act, landmark legislation that imposed sharp restrictions on shareholders suits. Two years ago, they pressured Congress to pass a law that bars state courts from hearing class-action lawsuits dealing with securities fraud. And earlier this year, their lawyers convinced a federal appeals court to interpret the Reform Act so strictly that future lawsuits against high-tech firms in Silicon Valley appear to have little chance of success. This is a story about how the high-tech industry has effectively shielded itself from the penalties for financial misconduct -- and what that means for investors and the future of the financial markets. It is about the SEC's heralded crackdown, and how it has failed to result in any significant legal actions curtailing financial accounting abuses in Silicon Valley. And it is about what appears to be a classic case of securities fraud that can't even get into court. THE CBT CASE, 1995 The sources were coming out of the woodwork, and San Francisco attorney Steven Sidener could hardly believe what they were telling him about a Redwood City software-maker called CBT Group. They mentioned a sales group staffed with so many ex-cons that it was known as C-Block. They described a shadow company that siphoned millions of dollars into directors' pockets. And they talked of altered credit card numbers and other tricks for faking sales. To Sidener it looked like a textbook case of securities fraud -- and three years ago, the shareholders' complaint describing phony accounting and extensive insider trading probably would have sailed into court. But Sidener was worried. The 1995 Reform Act had made it easier for judges to dismiss securities fraud cases. Still, the more Sidener learned about the actions of top CBT executives, the more he was convinced that if any case could meet the new standards, this was it. MAKING HIS MARK, 1997 CBT's chief executive officer is Gregory Priest, a former attorney at Wilson Sonsini Goodrich & Rosati, the most powerful law firm in Silicon Valley.

Raised in Los Gatos, Priest graduated from Princeton University and then Stanford Law School. In 1990, Priest joined Wilson Sonsini Goodrich & Rosati and quickly took to heart partner Larry Sonsini's mantra: Think like business people. Priest grew particularly close to one corporate client, CBT Group. On Nov. 22, 1995, just days after making partner at Wilson Sonsini, he left the firm to become CBT's chief financial officer. CBT made software that taught people how to use computer programs. After Priest arrived, it acquired six firms and extended its run of record revenues to 14 quarters. But Priest wanted to lead a company of his own, and he was growing restless. So in December 1997, he left CBT to become the chief executive officer of a company called Knowledge Well. Ostensibly, CBT and Knowledge Well were independent companies. But Priest remained a director of CBT, and William McCabe, who founded CBT and Knowledge Well, served as chairman of both companies. The companies' ties went even deeper, though, and those ties quickly led to trouble, according to former employees. Knowledge Well's goal was to create interactive computer programs for earning degrees from accredited universities. But many schools balked because professors would not grant credit for a course in which they played no active role, former employees say. When the company started to struggle, Priest, McCabe and other top CBT officials kept it afloat by diverting resources and millions of dollars from CBT to Knowledge Well, according to former employees and court records. The secret practice went on for months, propping up Knowledge Well even though the company was far from developing an accredited program for the market. And former employees at both companies say that was just one of many reasons they had to question the executives' conduct. ``They (Priest and McCabe) felt they were above the law,'' said a former CBT manager. ``They were so arrogant they felt they could get away with anything.'' PRESSURE TO PRODUCE, 1998 Meanwhile, CBT was headed for trouble of its own. Competitors were cutting prices. The sales staff complained that customers were rejecting CBT's standard three-year license deals because they didn't want to pay for software in advance. Sales were down, and renewals were off, according to court records. Although company executives assured the market that business was booming, they created an internal task force to investigate the company's financial troubles. They also boosted sales commissions to help increase revenues.

And on March 17, CBT announced that it would pay $150 million to acquire a Florida competitor called The ForeFront Group Inc. ForeFront sold its training software by phone with an aggressive staff of about 175 telemarketers, including one high-performing group nicknamed ``C-block'' or ``cell block'' because a number of ex-felons worked there, former employees say. When CBT took over, the pressure on ForeFront to increase sales grew even more intense. According to former employees and allegations in court documents, the sales people created phony invoices by altering customers' credit card numbers and sending orders from one fax machine to another in the same office. They booked revenues on products still being developed and on duplicate copies of sales invoices. They shipped defective products that eventually were returned, but they never recorded the returns. ``My whole thing was, I know CBT is here, and I know what they want, so let's give it to them,'' said a former employee. And while publicly proclaiming that CBT's sales were higher than ever, Priest, McCabe and other company officials began unloading millions of their CBT shares at near-peak values. On April 14, the company announced ``record'' revenues of $34 million for the first quarter of 1998, up 51 percent from the first quarter a year earlier. Two days later, Priest, McCabe and other company officials sold more than $13 million worth of company shares. On July 14, CBT announced another ``record'' quarter, and two days later, Priest sold more than $1.2 million worth of shares, missing the all-time high price by only 50 cents. Sidener was convinced that the timing of the officials' share sales was more than a coincidence. If that wasn't insider trading, he asked himself, what was? PLUNGING SHARE PRICES, 1998 For CBT, it all fell apart in a hurry. On Oct. 1, 1998, CBT announced that the company's revenues were down by one-third from the previous quarter. Within hours, CBT shares plummeted to 9§ from the July high of $63, costing investors, mutual funds and pension plans millions of dollars. On December 9, CBT announced that Priest had returned to the company as president and CEO. The next day, CBT announced that it had agreed to buy Knowledge Well.

Under the deal, Priest and McCabe received $60 million worth of CBT shares -- even though Knowledge Well still didn't have a product and had run up millions of dollars in losses in the two years since its founding. Priest defends the acquisition of Knowledge Well. ``Ninety-nine percent of the shareholders voted in favor of the deal, and when it was announced, CBT shares went up $4,'' he said. He contends that the sudden downturn in CBT's business ``surprised management'' -- even though McCabe had told employees during an October 1 phone call, ``We knew some time ago that we had problems.'' Priest also says there were ``no mistakes in booking revenue.'' But many shareholders were outraged. On their behalf, Sidener filed a class-action lawsuit accusing Priest, McCabe and other company executives of lying to shareholders and trading on inside information. Filed Nov. 24, 1998, the complaint was full of details -- more than enough, Sidener thought, to satisfy any court. But he knew that no matter how good his case against CBT, its fate depended on the outcome of another lawsuit before the U.S. Court of Appeals in San Francisco. A NEW ERA, 1999 The suit was called Silicon Graphics. It was the first class-action in California to test the Private Securities Litigation Reform Act of 1995, the federal law Silicon Valley executives helped push through Congress to limit securities fraud suits. Sidener worried that if the appeals court upheld a strict interpretation of the law, as high-tech industry lawyers were advocating, the case against CBT -- and scores of other companies accused of securities fraud – might never stand a chance. The Silicon Graphics case had been filed in January 1996 by plaintiffs' attorney Bill Lerach. It accused company executives of misleading investors by overstating the company's financial prospects. It also accused the executives of profiting from their alleged lies by selling millions of dollars worth of company stock. But in May 1997, U.S. District Judge Fern Smith dismissed the complaint, ruling that it didn't contain sufficient detail to satisfy the Reform Act's strict pleading standards. When Lerach appealed, he received help from a powerful ally: the Securities and Exchange Commission. In a supporting brief, the SEC argued that Smith's interpretation of the Reform Act was so severe that it would ``create a bar to meritorious claims.''

The agency urged the appellate court in San Francisco to overrule Smith, reminding the judges that their decision would ``have a significant impact . . . on the federal securities laws' protection of investors.'' But the appeals court refused. On July 3, 1999, a three-judge panel affirmed Smith's interpretation, breaking with two -- soon to be four -- other federal appeals courts and creating a standard that plaintiffs' lawyers and the SEC believed would be all but impossible for aggrieved shareholders to meet. ``It's a license for any judge with a predisposition against investors to dismiss any complaint,'' said Lerach, ``no matter how well pled, no matter how meritorious.'' The appeals court ruling worried Sidener, but he still felt optimistic about the CBT case. The appeals court had dismissed the Silicon Graphics complaint because shareholders had not provided enough detail to show that company executives knew their public statements were incorrect. But the perfect timing and the staggering volume of insider trades at CBT, the evidence of blatantly phony sales at ForeFront -- and McCabe's statement that ``we knew some time ago'' -- were more than enough, Sidener thought, to show that CBT's upbeat public statements were lies. ``These guys unloaded massive shares right before the bad news,'' Sidener said. ``That should have been enough by itself.'' But it wasn't. Less than three weeks after the appeals court decided the Silicon Graphics case, U.S. District Judge Ronald Whyte in San Jose dismissed the CBT complaint. ``A lot of firms,'' said Sidener, ``think this area (of the law) is finished.'' On October 4, Sidener refiled the CBT complaint anyway. AN OLD ENEMY, 1999 The decision in the Silicon Graphics case was only the latest defeat for Lerach, the lawyer whom high-tech executives had demonized as the ``prince of darkness.'' Less than two months earlier, he had suffered a humiliating personal blow when a Chicago jury ordered him and his law firm to pay an old nemesis $45 million. The events leading to the verdict began in 1989 when Lerach and other lawyers, representing thousands of bondholders, sued Charles Keating and Lincoln Savings & Loan for fraud.

A year later, Lerach added an economics consulting firm called Lexecon Inc. and one of its principals, Daniel Fischel, as defendants. Lerach says he sued them because they had once assured regulators that Lincoln was financially sound. In subsequent lawsuits against Apple Computer and other companies, Lerach and his colleagues attacked Fischel and undermined his testimony by calling him the ``crook'' who ``helped Lincoln Savings,'' according to court transcripts. The tactic proved so effective that many firms stopped hiring Fischel. So in 1992, Fischel sued Lerach and his law firm for slander and abuse of the legal process. The case went to trial last March, and from the outset of his testimony Lerach was his own worst enemy, quibbling over attorneys' questions and often annoying the judge. But the case may have turned on Lerach's own words: Colleagues testified that Lerach said he had added Fischel to the Lincoln case ``because I want to put the little f---er out of business.'' Although the judgment was $45 million, Lerach and his firm quickly settled for $50 million to avoid paying even more if the jury assessed punitive damages. It was a devastating comeuppance for one of corporate America's greatest enemies. Today, in his San Diego office, he can barely discuss the case. He sits silently for several moments, then tersely comments, ``It was an acrimonious and hard-fought trial.'' The only thing that may have hurt Lerach as much as the amount of the award was the unconcealed delight the nation's corporate executives took in his having to pay it. And no one was more amused than Alan Shugart, founder and former CEO of Seagate Technology -- and the focus of four Lerach lawsuits. ``It couldn't happen to a nicer guy,'' he said. THE SECURITIES AND EXCHANGE COMMISSION, 1998 Lynn Turner knew exactly what was going on. And that was what SEC Chairman Arthur Levitt was looking for in July 1998 when he brought Turner to the agency's Washington headquarters. In recent years, the job of protecting investors from securities fraud had fallen almost solely on the SEC.

The Reform Act and a second bill passed by Congress in 1998 barring shareholder suits in state courts effectively shielded corporate executives from any legal action brought by investors. And federal prosecutors had filed only a handful of criminal cases against Silicon Valley executives. Turner, a former partner at Coopers & Lybrand in Denver, one of the nation's largest accounting firms, had specialized in high-tech auditing for 15 years. He then served two years as chief financial officer of Symbios Logic Inc., a Denver semiconductor-chip maker. When Levitt tapped him to become the SEC's chief accountant and lead the agency's campaign against accounting abuses, he was more than prepared. ``When I got here,'' he said, ``I knew exactly where to look. I knew exactly how people were playing the game and what was happening.'' Turner was particularly concerned about the ``astronomical'' number of research and development write-offs that high-tech companies were taking when they acquired another company. The write-offs allow a company to set a value on the research and development in process at the companies it acquires -- and then to deduct it from the bottom line as a one-time charge. And Turner knew that many companies were grossly inflating the value of the ``in process'' R&D they acquired. One company, for example, claimed R&D write-offs for products that were already in the marketplace winning awards. Many write-offs take place in a gray zone, where accounting rules are flexible. But in some cases, executives were using the write-offs to boost future earnings and mask their company's fundamental weaknesses. ``Even the (Wall Street) analysts can't see what's really going on,'' Turner said. And the questionable accounting practices were spreading. ``I've had one major player in Silicon Valley tell me that in the last month, based on what he has seen, he doesn't know if there has been a single decent (R&D) valuation in the valley,'' Turner said. So Turner and other top SEC officials quickly singled out suspect companies, including AOL, Compaq, and other prominent firms. In January, the SEC sent letters to 150 companies, notifying them that the agency intended to review the billions of dollars they had claimed as R&D write-offs, as well as other suspicious accounting charges.

It wasn't long before one of those letters arrived at the Santa Clara offices of a security software company called Network Associates. ROLLING THE DICE Network Associates was run by William Larson, a hard-charging executive who had raced cars and created a business motto: ``Drive fast -- take chances.'' The Massachusetts native graduated from Harvard and earned a master's at Wharton School of Business by the age of 22. Three years later he left Stanford with a law degree. He worked as a top sales executive at Sun Microsystems until 1993, when he left to run McAffee Associates, a small antivirus software firm. At the time, the 4-year-old company was so tiny it was distributing its software on computer bulletin boards, trusting its users to mail in the sales payments. Its annual sales totaled $20 million. Its staff numbered 40. But Larson had bigger plans. In 1994, he bought a New Jersey network tools developer, then a company called Automated Design Systems. He spent $91.8 million to acquire another company the next year. In 1996, Larson bought three companies, then eight more the next year. But none of those acquisitions compared with the blockbuster deal Larson pulled off in October 1997 -- a $1.3billion bid for Network General, a Menlo Park software company. The company that emerged -- with 1,600 employees and $600 million in revenue -- was dubbed Network Associates, and Larson was its chairman and CEO. ``I could have left McAffee alone,'' said Larson. ``But we put that all back on the craps table and said: `Let's roll the dice again and go to the next level.' '' And Larson's buying binge was far from over. The next year, he spent nearly $1.5 billion to acquire 10 more companies. Many of the purchases were paid for with Network Associates' stock, but others were made with cash, enabling the company to take multimillion-dollar R&D write-offs. The strategy paid off handsomely. Network Associates was soon one of the leading software companies in the world, with each purchase broadening the company's stable of encryption, antivirus, network management and security products and services. And its stock price soared. RUMOR AND DENIAL

In late 1998, Internet chat lines buzzed with rumors that the rapid-fire purchases, inflated acquisition charges and other questionable accounting practices concealed weaknesses in Network Associates' overall performance. In early September, when the company's stock began to slump, a furious Larson promptly called a phone conference with financial analysts to defend the company. ``The numbers in our write-offs are squeaky clean,'' he insisted. ``Those who believe (in Network Associates) have been richly rewarded in the past,'' he said, referring to the value of the company's stock, which had multiplied nearly 15 times over five years. ``And they will be richly rewarded in the future. And I'm one of them.'' The next day, Network Associates' stock price shot up 9.2 percent. Later that month, the company made its last acquisition, buying a diagnostic software company called CyberMedia for $131 million -- and writing off an astounding 93 percent of the acquisition's cost as in-process research and development. THE BIG BANG, 1999 For Larson, it was the ultimate marketing moment. On April 5, future Hall of Famer Roger Clemens was set to take on the Oakland A's at what is now called Network Associates Coliseum. The company had paid $5.8 million for the right to put its name on Alameda County's outdoor sports stadium, and Larson had invited 500 people to the Arena's Plaza Club next door for a special pre-game event. To herald the occasion, he had issued 13 different press releases and placed full-page ads in newspapers. The 42-year-old executive beamed as he gazed out at his guests, including Oakland Mayor Jerry Brown and representatives from Microsoft and Hewlett Packard and Sun Microsystems. After introducing a new suite of management and security products, Larson christened the stadium ``The Net.'' ``We are opening the 1999 baseball season with a bang,'' he said. But for tens of thousands of Network Associates shareholders, the big bang would come the next day. On April 6, word got out that the SEC's review would force Network Associates to restate its 1997 and 1998 earnings. That same day, 21 million shares changed hands -- seven times the stock's daily average. The company's stock value plunged 25 percent, costing shareholders hundreds of millions of dollars. In a hastily arranged telephone conference after the markets closed, Larson acknowledged that Network Associates had settled with the Securities and Exchange Commission. The

company had agreed to erase more than $214 million in R&D charges written off during the past two years and to subtract the charges from future earnings instead. For five straight years, Network Associates had reported steady earnings growth and had exceeded Wall Street estimates. Now Larson reported that the company's profits for the first quarter of this year would fall 38 percent below earlier projections. The news would only get worse. Two weeks later, Larson announced that the company's revenues in the second quarter would fall to almost zero -- a drop of nearly a quarter of a billion dollars -- as the company cleared out unsold inventory. Larson blamed the sales slowdown on customer concerns about Y2K and longer sales cycles for the company's complex new products. The next day, the company's stock dropped an additional 27 percent. From a high of $66 on January 1, Networks Associates' stock price had plummeted to $10 a share. In just four months, the company's market value had plunged from $9.2 billion to $1.4 billion. Security analysts -- most of whom had touted Network Associates as a strong growth stock -- were in shock. Larson had lost all credibility with Wall Street. On Internet message boards, Larson was vilified as a ``snake oil salesman,'' and shareholders soon filed dozens of lawsuits accusing him of securities fraud. Larson would later say the criticism ``only makes me more motivated.'' And he dismissed the SEC investigation with a wave of his hand. ``It didn't change our revenue numbers,'' he said. ``It just changed the accounting.'' THE CRACKDOWN, 1999 It was not a coincidence. On September 28, exactly a year after Levitt called for an aggressive crackdown on improper financial reporting, the SEC proudly announced that it had taken enforcement actions against 68 people at 15 public companies, mostly for schemes that falsely inflated the value of their companies' revenues. Only 11 were chief executive officers. The best-known was Fran Tarkenton, a former NFL quarterback and television personality. Tarkenton, head of an Atlanta software company, had agreed to pay a $100,000 fine but admitted no wrongdoing.

The rest worked at small companies in Illinois, Florida, Virginia and elsewhere. None of the executives or companies were from Silicon Valley. During the past year, Network Associates and more than 40 other companies have reversed write-offs and restated earnings as a result of pressure from the SEC. But the agency has filed no charges against Network Associates or the other 149 companies targeted in January. The agency declines to say whether any enforcement action is planned. Today, more than a year after Levitt's dramatic speech at New York University, the SEC says the 150 letters it sent out in January were intended for guidance only. ``No agency can eliminate all fraud,'' said Levitt. ``It's a cultural change we are seeking.'' THE DROP ZONE, 1999 It had been a tough year for Network Associates, but Larson had done well for himself. Last year, when his company's stock was soaring, he cashed in $20 million in stock options. He also took home salary and bonuses worth nearly $1 million. His total compensation package ranked him among the 10 highest-paid executives in Silicon Valley for the third year. But now it was time to face the shareholders. At the company's annual meeting at the Marriott Hotel in Santa Clara, Larson told the investors that he often looks out the window of his 11th-floor office at Great America's ``Drop Zone,'' a 224-foot-high thrill ride that offers a terrifying 3 G's drop. When the bottom fell out of Network Associates' stock, he joked, ``that's what it felt like for me all the way down.'' Only a few shareholders laughed. EPILOGUE On a muggy day in early June, a group of high-tech executives made the rounds in Washington, D.C., led by pre-eminent Silicon Valley venture capitalist John Doerr, the financial midwife to Cypress, Netscape, Sun Microsystems and many other high-tech enterprises. Earlier that day, the group met with Vice President Al Gore's staff and House Democrats to lobby for a laundry list of legislation, ranging from bills to preserve research and development write-offs to proposals for shielding high-tech companies from lawsuits involving Y2K breakdowns. Then the executives sat down with nine Senate Democrats in the Capitol Hill conference room of Sen. John Kerry, D-Mass. Most of the senators knew Doerr, a longtime adviser to Gore and a

major contributor to Democratic Party candidates. But they were unprepared for the blistering tirade he delivered. In May, Kerry had drafted a bill to protect businesses from frivolous Y2K lawsuits while preserving the right to sue for damages caused by Y2K computer problems. But Kerry's bill, which failed to pass, did not go nearly far enough to satisfy Doerr and the other executives. They wanted the senators to support a bill by Sen. John McCain, R-Ariz., that provided much stronger legal protection for the technology industry. Doerr made it unmistakably clear that they were ``either with Silicon Valley or with the trial lawyers,'' said one staff member familiar with the meeting. The senators were stunned, he said. ``Don't forget, these were Democrats who had been there for Silicon Valley, who had sponsored all kinds of legislation favorable to high-tech.'' Another staff member was equally astonished. ``Anyone else would have been ushered out of the room,'' he said. ``But high-tech is now so powerful they can get away with it.'' Three weeks later, on July 1, McCain's Y2K bill -- with some consumer protections added at the last minute by the Clinton administration -- passed by an 81-to-18 vote in the Senate. The margin was even greater in the House, where the bill passed 404 to 24. -- -- -- The same day, Texas Gov. George W. Bush delivered a rousing speech before 500 high-tech executives who had paid $1,000 each for the event in Palo Alto. It was Bush's first official visit to Silicon Valley as a presidential candidate, and he touted his support for a roster of issues close to his listeners' hearts. Bush said that he would ``take the side of innovation over litigation every single time,'' and that government regulation should not be used to strangle initiative in the high-tech industry. Silicon Valley, he proclaimed, ``speaks to the spirit of America . . . and the ultimate American dream.'' -- -- -- Across the country, in Lafayette, LA, Charles Pollock lay in bed, unable to speak as a result of a stroke he suffered five years ago.

In the early 1980s, the 71-year-old former petroleum engineer had invested in Seagate Technology, trying to prepare for his retirement. When the stock crashed in 1984, Pollock lost $544.99. The loss might not seem like a lot, said his wife Diane, who has been forced sell some of the family's possessions to meet her husband's mounting medical bills. ``But to us back then it was,'' she said. ``We really need the money now.'' REFORM ACT The Private Securities Litigation Reform Act of 1995 was designed to limit class-action lawsuits for federal securities fraud. It makes filing the suits more difficult by requiring shareholders to show in specific detail that company officers intended to defraud investors. Federal courts in California have interpreted the act to require more detail than other courts have. The Reform Act requires courts to choose a class action's lead plaintiff, usually a shareholder that owns the most stock. The act also shields companies from suits over statements about the future and makes each defendant liable only for its share of wrongdoing. THE SECURITIES AND EXCHANGE COMMISSION The SEC's San Francisco office has filed nine financial fraud and insider trading cases this year. Two cases allege misconduct by auditors and executives in the California Micro Devices case. In two other cases, Insignia Solutions, a Fremont software firm, was accused of overstating revenue by $2.4 million, and WIZ Technology of San Juan Capistrano, was charged with inflating income and revenue by $1.6 million. In four cases, the SEC sought fines against individuals who profited from stock trades based on inside information. The agency also took action against a Palo Alto man for fraudulent sales of securities over the Internet.